News
About Us
Research
Team
Publications
Career
xchange
Contact
  • EN DE

    • Publications

    We publish our research papers at a wide spectrum of venues. In addition, we aim at releasing our research results in the form of open source software projects and datasets.

    emergenCITY Policy Paper Series

    Information and communication technology (ICT) plays a crucial role in the response to and ongoing management of crises. Policy Paper 1 presents reasons for the increasing fragility and decreasing resilience of ICT and suggests recommendations for increasing and sustainably ensuring the resilience of current and future ICT. In the view of the emergenCITY researchers, ICT can only be described as resilient if it can maintain an acceptable minimum or substitute functionality despite significant impairments and if it is equipped for a rapid return to normal behavior.
    Also available at emergencity.de/s/pp1.

    Download the paper

    Human or technical failure, natural disasters, pandemics, cyber or terrorist attacks can lead to a nationwide power outage in Germany that lasts longer than 24 hours. Cities are then faced with the challenge of having to react in order to successfully bridge the crisis. Policy Paper 2 examines the measures taken by the local disaster control offices of major German cities in response to this scenario. In addition, the emergenCITY researchers show further options for action on how cities can be better prepared for a long-lasting supra-regional power outage in the future.
    Also available at emergencity.de/s/pp2.

    Download the paper

    Current disaster incidents, such as the heavy rainfall event in the Ahr Valley in summer 2021, are opening up new debates on how cooperation between emergency services in German civil protection and disaster management can be better coordinated. The policy paper 3 analyzes current weaknesses in disaster management. The emergenCITY researchers also provide recommendations on how coordination processes between those involved in disaster management can be improved in direct response to and management of a crisis situation.
    Also available at emergencity.de/s/pp3.

    Download the paper

    Successful urban disaster management requires cooperation between authorities, organizations, companies and the population. Participation plays a central role here, as the early involvement of stakeholders increases the quality and acceptance of decisions. Policy Paper 4 shows different levels of participation and the ideal-typical sequence of a cooperative participation process for the development of a crisis management plan. The emergenCITY researchers emphasize that co-creation processes are crucial for increasing the resilience of cities and overcoming crises.
    Also available at emergencity.de/s/pp4.

    Download the paper

    The dependency of agricultural businesses on communication and energy infrastructure is increasing, but the impact of this current trend on the resilience of the sector remains unclear. Policy Paper 5 takes a critical perspective on specific security risks in the ongoing digitalization and increasing importance of data in agriculture. The emergenCITY researchers emphasize that improved security and resilience to infrastructure failures and the protection of data are essential for digitalization to sustainably improve the situation of farms.
    Also available at emergencity.de/s/pp5.

    Download the paper

    Scientific Publications

    Filter by Type

    Filter by Year

    Sort by Date

      March 2025  Signal Processing Article

    High-Dimensional False Discovery Rate Control for Dependent Variables

    Jasin Machkour, Michael Muma, Daniel P. Palomar

    PDF BibTeX DOI: 10.1016/j.sigpro.2025.109990

    Abstract
    Algorithms that ensure reproducible findings from large-scale, high-dimensional data are pivotal in numerous signal processing applications. In recent years, multivariate false discovery rate (FDR) controlling methods have emerged, providing guarantees even in high-dimensional settings where the number of variables surpasses the number of samples. However, these methods often fail to reliably control the FDR in the presence of highly dependent variable groups, a common characteristic in fields such as genomics and finance. To tackle this critical issue, we introduce a novel framework that accounts for general dependency structures. Our proposed dependency-aware T-Rex selector integrates hierarchical graphical models within the T-Rex framework to effectively harness the dependency structure among variables. Leveraging martingale theory, we prove that our variable penalization mechanism ensures FDR control. We further generalize the FDR-controlling framework by stating and proving a clear condition necessary for designing both graphical and non-graphical models that capture dependencies. Additionally, we formulate a fully integrated optimal calibration algorithm that concurrently determines the parameters of the graphical model and the T-Rex framework, such that the FDR is controlled while maximizing the number of selected variables. Numerical experiments and a breast cancer survival analysis use-case demonstrate that the proposed method is the only one among the state-of-the-art benchmark methods that controls the FDR and reliably detects genes that have been previously identified to be related to breast cancer. An open-source implementation is available within the R package TRexSelector on CRAN.

      February 2025  IEEE Transactions on Wireless Communications Article

    RISnet: A Domain-Knowledge Driven Neural Network Architecture for RIS Optimization with Mutual Coupling and Partial CSI

    Bile Peng, Karl-Ludwig Besser, Shanpu Shen, Finn Siegismund-Poschmann, Ramprasad Raghunath, Daniel Mittleman, Vahid Jamali, Eduard A. Jorswieck

    BibTeX DOI: 10.1109/TWC.2025.3536178

    Abstract
    Space-division multiple access (SDMA) plays an important role in modern wireless communications. Its performance depends on the channel properties, which can be improved by reconfigurable intelligent surfaces (RISs). In this work, we jointly optimize SDMA precoding at the base station (BS) and RIS configuration. We tackle difficulties of mutual coupling between RIS elements, scalability to more than 1000 RIS elements, and high requirement for channel estimation. We first derive an RIS-assisted channel model considering mutual coupling, then propose an unsupervised machine learning (ML) approach to optimize the RIS with a dedicated neural network (NN) architecture RISnet, which has good scalability, desired permutation-invariance, and a low requirement for channel estimation. Moreover, we leverage existing high-performance analytical precoding scheme to propose a hybrid solution of ML-enabled RIS configuration and analytical precoding at BS. More generally, this work is an early contribution to combine ML technique and domain knowledge in communication for NN architecture design. Compared to generic ML, the problem-specific ML can achieve higher performance, lower complexity and permutation-invariance.

      January 2025  Technische Universität Darmstadt Darmstadt Thesis

    Distribution Grid Restoration using Microgrids: Interdependencies between Power, Water, and Communication Networks

    Martin Robison Pietsch

    PDF BibTeX DOI: 10.26083/tuprints-00029045

    Abstract
    The resilience of urban infrastructure systems—particularly power, water, and communication networks—is crucial for maintaining society’s functions during and after crises such as natural disasters or cyber-attacks. Traditional approaches to dealing with disasters often treat power, water, and communication networks independently, overlooking dependencies and synergies that could be critical during crises. In this dissertation we propose new methods to improve the recovery process in this context. All those methods have three strategies in common, which are applied to different problems: The dynamic formation of decentral microgrids (MGs), the integration of renewable energy sources and the mutual benefit of the interdependencies of electricity, water and communication networks in the restoration process. We start by integrating water networks with electrical MGs. Using the backup generators of the pumps of the water network to initiate local MGs, we can use renewable energy sources in the surrounding neighborhood to reduce fuel consumption during a crisis. Optimizing additional valves and switches in each network further allows us to evenly distribute and schedule the limited available resources. After ensuring the local restoration of individual MGs, we address how merging them can further enhance overall system stability. We therefore develop a novel framework to merge individual MGs, considering the explicit need for an active communication link between the MGs generators as a second contribution. Here, the communication network depends on a functioning power supply to keep a connection online. Merging MGs enhances the operation stability and allows the restoration of larger electrical loads within the network. In case of damaged communication hardware, we extend our approach to cover gaps in the communication network via unmanned aerial vehicles (UAVs) and optimize their placement considering close renewable battery charging stations. Strategic placement of UAVs ensures continuous communication capability, which is vital for coordinated restoration efforts and efficient management of distributed resources. However, as the optimization of the UAV placement can lead to large combinatorial problems, we increase computation speed by introducing the feasibility query algorithm (FQA), excluding infeasible solutions beforehand. Finally, we propose a community-power-sharing model, allowing distributed renewable energy resources to support nearby communication infrastructure via MGs using the distribution grid during outages, ensuring critical service continuity. We demonstrate this approach on two smart buildings at the campus Lichtwiese in Darmstadt and show how power sharing can be implemented in a real-life setup. All problems are formulated as mixed-integer linear programs (MILP) and represent different ways to model the interdependencies between power, water, and communication networks. This allows for explicit decision-making and scheduling of the restoration process. Overall, this dissertation aims to enhance urban infrastructure resilience by leveraging MGs and renewable energy resources to integrate and restore interdependent power, water, and communication networks. By introducing integrated solutions that leverage these interdependencies and exploit renewable energies, the research not only enhances the resilience of individual networks but promotes a holistic and mutually beneficial approach to urban crisis management. Our findings offer practical implications for policy making, for shaping economic incentives and the design of resilient urban energy systems and point towards a range of opportunities for future research and further practical implementations.

      January 2025  Signal Processing Article

    The Terminating-Random Experiments Selector: Fast High-Dimensional Variable Selection with False Discovery Rate Control

    Jasin Machkour, Michael Muma, Daniel P. Palomar

    PDF BibTeX DOI: 10.1016/j.sigpro.2025.109894

    Abstract
    We propose the Terminating-Random Experiments (T-Rex) selector, a fast variable selection method for high-dimensional data. The T-Rex selector controls a user-defined target false discovery rate (FDR) while maximizing the number of selected variables. This is achieved by fusing the solutions of multiple early terminated random experiments. The experiments are conducted on a combination of the original predictors and multiple sets of randomly generated dummy predictors. A finite sample proof based on martingale theory for the FDR control property is provided. Numerical simulations confirm that the FDR is controlled at the target level while allowing for a high power. We prove under mild conditions that the dummies can be sampled from any univariate probability distribution with finite expectation and variance. The computational complexity of the proposed method is linear in the number of variables. The T-Rex selector outperforms state-of-the-art methods for FDR control on a simulated genome-wide association study (GWAS), while its sequential computation time is more than two orders of magnitude lower than that of the strongest benchmark methods. The open source R package TRexSelector containing the implementation of the T-Rex selector is available on CRAN.

      2025  Sustainability Analytics and Modeling Article

    Improving the resilience of socio-technical urban critical infrastructures with digital twins: challenges, concepts, and modeling

    Tobias Gebhard, Bernhard J. Sattler, Jonas Gunkel, Marco Marquard, Andrea Tundis

    PDF BibTeX DOI: 10.1016/j.samod.2024.100036

    Abstract
    The increasing number of crises, including natural disasters and military conflicts, underscores the importance of resilient critical infrastructures (CIs), especially for urban areas. However, current approaches for CI modeling, monitoring, and resilience assessment are lacking a holistic view of cities as complex, interconnected, and socio-technical systems. This paper explores the application of the Digital Twin (DT) concept as a promising tool to assess and improve the resilience of urban CIs in light of various hazards. DTs are virtual real-time representations of a physical system that can be used to perform real-time analysis, simulate what-if scenarios, and provide decision support, during crises and normal operations. To this end, we identify and discuss key challenges for the development of Urban Digital Twins (UDTs), including data management, technical and social modeling of CIs, integrated CI co-simulations, model validation, and resilience assessment. To address the complex nature of urban areas as systems-of-systems, we present overarching modeling concepts by considering CI interdependencies and socio-technical interactions, resulting in the concept of the Socio-technical Digital Twin (STDT). Beside incorporating agent-based modeling, we discuss the issue of demand synchronization and propose the concepts of model selection and model transfer to facilitate the modeling process for UDTs. Furthermore, a multi-layered modeling framework for interdependent urban CIs is presented, where the proposed concepts are integrated and an overview and discussion of the technical and social modeling of CIs is provided, with a particular focus on the power, water, and transportation domain. Finally, we deal with the quantitative resilience assessment for interconnected CIs and discuss ways of integrating these methodologies in DTs. Our approach frames CIs as socio-technical systems and integrates the human perspective into the modeling process and resilience assessment. The presented modeling framework can be used to simulate various scenarios for analyzing their consequences in advance and measuring resilience in a holistic context. Moreover, the proposed concepts and modeling approaches can support future developments towards UDTs for crisis management.

      December 2024  IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN) Conference

    State-aware Subscriber Steering in Fiber Access Networks for Improved Resilience

    Fridolin Siegmund, Philip Jonas Franz, Bjoern Nagel, Maik Rüder, Lisa Wernet, Matthias Hollick, Ralf Kundel

    PDF BibTeX DOI: 10.1109/NFV-SDN61811.2024.10807483

    Abstract
    Network softwarization is currently emerging in home and enterprise Internet access networks. Still, the advantages of programmable hardware nodes and virtualized network functions in the access network as resilience enablers should be researched more. Access networks are typically built in a two-layer tree topology with sparse redundancy and limited adaptability. To overcome this, this work investigates how softwarization with redundancy can achieve resilience improvements in such networks. We present an architecture and prototypical implementation of a fiber-based access network and subscriber termination, with a redundant and adaptive design throughout the whole path from subscribers to the Internet. By proactively migrating states and steering subscriber flows, we reached total average downtimes of around 156 milliseconds to restore Internet access after disruption in the case of one subscriber. In comparison, it takes hours to restore Internet access in case of an outage with zero redundancy. Our solution is cost-effective by exploiting the architectural advances of Software Defined Networking, allowing the use of virtualized network functions and hardware-based functions simultaneously.

      December 2024  International Conference on Intelligent Robots and Systems (IROS 2024) Conference

    Efficient Dynamic LiDAR Odometry for Mobile Robots with Structured Point Clouds

    Jonathan Lichtenfeld, Kevin Daun, Oskar von Stryk

    PDF BibTeX DOI: 10.1109/IROS58592.2024.10802045

    Abstract
    We propose a real-time dynamic LiDAR odometry pipeline for mobile robots in Urban Search and Rescue (USAR) scenarios. Existing approaches to dynamic object detection often rely on pretrained learned networks or computationally expensive volumetric maps. To enhance efficiency on computationally limited robots, we reuse data between the odometry and detection module. Utilizing a range image segmentation technique and a novel residual-based heuristic, our method distinguishes dynamic from static objects before integrating them into the point cloud map. The approach demonstrates robust object tracking and improved map accuracy in environments with numerous dynamic objects. Even highly non-rigid objects, such as running humans, are accurately detected at point level without prior downsampling of the point cloud and hence, without loss of information. Evaluation on simulated and real-world data validates its computational efficiency. Compared to a state-of-the-art volumetric method, our approach shows comparable detection performance at a fraction of the processing time, adding only 14 ms to the odometry module for dynamic object detection and tracking. The implementation and a new real-world dataset are available as open-source for further research.

      December 2024  18th ACM Workshop on Wireless Network Testbeds, Experimental evaluation & Characterization 2024 Conference

    Beyond Sensing: A High-Performance Software-Defined LoRa Gateway

    Vincenz Mechler, Matthias Hollick, Bastian Bloessl

    PDF BibTeX DOI: 10.1145/3636534.3697317

    Abstract
    LoRa is about to become the standard for Low-Power Wide-Area Networks (LPWANs), being well suited for many Internet of Things (IoT) and sensor network applications. The success of the technology sparked interest to adopt LoRa for other, more challenging use-cases in industrial automation or control of cyber-physical systems. Regular LoRa gateways are, however, limited in their number of parallel demodulation paths and restricted to a single network. To overcome this limitation, we implement a software-defined, multichannel LoRa receiver that is able to receive all common spreading factors and uplink channel combinations of the LoRaWAN EU868 frequency plan. In addition, our receiver does not rely on hard-coded sync words, enabling a mechanism similar to monitor mode in WLAN networks. Experimental evaluation using Software-Defined Radios (SDRs) confirms superior performance compared to commercial LoRa gateways.

      December 2024  14th IEEE Global Humanitarian Technology Conference (GHTC2024) Conference

    On the Feasibility of Digital VHF Communications for Crisis Scenarios

    Michi Hermann, Bastian Bloessl

    BibTeX DOI: 10.1109/GHTC62424.2024.10771549

    Abstract
    Communication is key, especially during a crisis. In case of larger incidents, existing infrastructure often faces outages, making it difficult - if not impossible - to communicate. While first responders have communication solutions for these cases, civilians remain disconnected. To avoid such stressful situations, we study Rattlegram, a novel communication solution that uses acoustic coupling of smartphones with VHF/UHF handheld radios to transmit digital information. We compare it to existing protocols for the VHF/UHF band and evaluate its performance in simulations and field tests. We demonstrate its applicability, showing that even a cheap handheld radio can be used to establish communication up to a distance of over 18km in rural and close to 1 km in urban environments. The results highlight that Rattlegram is a viable technology that can be the base for affordable, infrastructure-less emergency communication for the masses.

      December 2024  Results in Engineering Article

    Integrated risk analysis for urban flooding under changing climates

    Wenyu Yang, Ziyong Zhao, Liping Pan, Ruifei Li, Shixue Wu, Pei Hua, Haijun Wang, Britta Schmalz, Peter Krebs, Jin Zhang

    BibTeX DOI: 10.1016/j.rineng.2024.103243

    Abstract
    Urban flooding poses significant threats to human lives and urban development worldwide, while the impact of climate change on urban flooding remains unclear. To systematically analyze the variabilities of hydrological patterns and urban flooding under changing climates, this study coupled the downscaled general circulation model (GCM) projection with hydrologic-hydraulic modeling, with consideration of both the low-greenhouse gas (GHG) emission scenario of SSP1–2.6 and the high-GHG emission scenario of SSP5–8.5. Results demonstrated that the original GCM projection effectively captured the changing trend of rainfall patterns in the given area, with an overestimation of rainfall peaks. The GCM downscaling through the k-nearest neighbors (kNN)-based analog method significantly improved the model accuracy. Scenario analysis indicated that climate change significantly affected the regional hydrology, with the precipitation, surface runoff, and floods increasing by a maximum value of 17.10%, 12.66%, and 63.26%, respectively. The interannual comparison demonstrated that the temporal variability in precipitation and flood intensified with the increase in GHG emissions during 2025–2100, suggesting the uncertainty of long-term climate forecasts. According to flood risk analysis, long-term and short-term floods exhibit varied changing trends across climate scenarios, despite the strong correlations between precipitation and runoff, implying the complexity of flood generation mechanisms. The methods and findings herein provide valuable insights for policymakers and practitioners, to cope with the increasing urban flood risk within the context of evolving environment.

      December 2024  IEEE Transactions on Wireless Communications Article

    Physical-Layer Privacy via Randomized Beamforming Against Adversarial Wi-Fi Sensing: Analysis, Implementation, and Evaluation

    Marco Cominelli, Shaghayegh Shahcheraghi, Jakob Link, Matthias Hollick, Federico Cerutti, Francesco Gringoli, Arash Asadi

    BibTeX DOI: 10.1109/TWC.2024.3485477

    Abstract
    Wi-Fi sensing applications have achieved remarkable results over the last decade, offering accurate device-free localization and gesture recognition capabilities. Indeed, Wi-Fi sensing has quickly become a critical field of research for future communication systems under the paradigm known as joint communication and sensing. However, device-free wireless sensing can also be exploited for malign purposes against unaware victims, and the omnipresence of Wi-Fi transceivers poses a significant threat to people’s privacy. Therefore, it is essential to develop functional solutions that can effectively thwart wireless sensing. All the current attempts to hinder illegitimate wireless sensing rely on specialized hardware deployed in the environment, but their cost and complexity can undermine widespread deployment. In this paper, we explore the possibility of using native capabilities of Wi-Fi systems, namely beamforming, to thwart wireless sensing. To this end, we propose for the first time a solution that enables complete control over the beamforming in commercial Wi-Fi devices. On top of that, we build BeamDancer, which randomizes beamforming vectors to inhibit channel fingerprinting. We empirically demonstrate the effectiveness of the proposed solution against three different wireless sensing techniques, both data-driven and model-based, while preserving almost entirely the legitimate Wi-Fi traffic at the same time.

      November 2024  Technische Universität Darmstadt Darmstadt Thesis

    Development of Fast Machine Learning Algorithms for False Discovery Rate Control in Large-Scale High-Dimensional Data

    Jasin Machkour

    PDF BibTeX DOI: 10.26083/tuprints-00028231

    Abstract
    This dissertation develops false discovery rate (FDR) controlling machine learning algorithms for large-scale high-dimensional data. Ensuring the reproducibility of discoveries based on high-dimensional data is pivotal in numerous applications. The developed algorithms perform fast variable selection tasks in large-scale high-dimensional settings where the number of variables may be much larger than the number of samples. This includes large-scale data with up to millions of variables such as genome-wide association studies (GWAS). Theoretical finite sample FDR-control guarantees based on martingale theory have been established proving the trustworthiness of the developed methods. The practical open-source R software packages TRexSelector and tlars, which implement the proposed algorithms, have been published on the Comprehensive R Archive Network (CRAN). Extensive numerical experiments and real-world problems in biomedical and financial engineering demonstrate the performance in challenging use-cases. The first three main parts of this dissertation present the methodological and theoretical contributions, while the fourth main part contains the practical contributions. The first main part (Chapter 3) is dedicated to the Terminating-Random Experiments (T-Rex) selector, a new fast variable selection framework for high-dimensional data. The proposed T-Rex selector controls a user-defined target FDR while maximizing the number of selected variables. This is achieved by fusing the solutions of multiple early terminated random experiments. The experiments are conducted on a combination of the candidate variables and multiple independent sets of randomly generated dummy variables. A finite sample proof of the FDR control property is provided using martingale theory. The computational complexity of the T-Rex selector grows linearly with the number of candidate variables. Furthermore, its computation time is more than two orders of magnitude faster compared to state-of-the-art benchmark methods in large-scale data settings. Therefore, the T-Rex selector scales to millions of candidate variables in a reasonable computation time. An important use-case of the T-Rex selector is determining reproducible associations between phenotypes and genotypes in GWAS, which is imperative in personalized medicine and drug discovery. The second main part (Chapter 4) concerns dependency-aware FDR-controlling algorithms for large-scale high-dimensional data. In many biomedical and financial applications, the high-dimensional data sets often contain highly correlated candidate variables (e.g., gene expression data and stock returns). For such applications, the dependency-aware T-Rex (T-Rex+DA) framework has been developed. It extends the ordinary T-Rex framework by accounting for dependency structures among the candidate variables. This is achieved by integrating graphical models within the T-Rex framework, which allows to effectively harness the dependency structure among variables and to develop variable penalization mechanisms that guarantee FDR control. In the third main part (Chapter 5), algorithms for joint grouped variable selection and FDR control are proposed. This approach to tackling the challenges resulting from the presence of groups of highly dependent variables in the data is different to the more conservative variable penalization approach that has been developed in the second part of this dissertation. That is, instead of finding the few true active variables among groups of highly correlated variables, the goal is to select all groups of highly correlated variables that contain at least one true active variable. In genomics research, especially for GWAS, grouped variable selection approaches are highly relevant, since one is not interested in identifying a few single-nucleotide polymorphisms (SNPs) that are associated with a disease of interest but rather the entire groups of correlated SNPs that point to relevant locations on the genome. The fourth main part of this dissertation (Chapters 6 and 7) demonstrates the application of the developed methods to practical problems in biomedical engineering as well as financial engineering. The biomedical applications include (i) a semi-real-world GWAS, (ii) a human immunodeficiency virus type 1 (HIV-1) data set with associated drug resistance measurements, and (iii) a breast cancer data set with associated survival times of the patients. The financial engineering applications include (i) accurately tracking the S&P 500 index using a quarterly updated and rebalanced tracking portfolio that consists of few stocks and (ii) a factor analysis of S&P 500 stock returns. The common challenge of all considered applications lies in detecting the few true active variables (i.e., SNPs, mutations, genes, stocks) among many non-active variables in, among other things, large-scale high-dimensional settings. Summarizing, this dissertation develops and analyses new fast and scalable machine learning algorithms with provable FDR-control guarantees for variable selection tasks in large-scale high-dimensional data. The developed algorithms and the associated open-source software packages have enabled making reproducible discoveries in various real-world applications ranging from biomedical to financial engineering.

      November 2024  27th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2024) Conference

    PairSonic: Helping Groups Securely Exchange Contact Information

    Florentin Putz, Steffen Haesler, Thomas Völkl, Maximilian Gehring, Nils Rollshausen, Matthias Hollick

    PDF BibTeX DOI: 10.1145/3678884.3681818

    Abstract
    Securely exchanging contact information is essential for establishing trustworthy communication channels that facilitate effective online collaboration. However, current methods are neither user-friendly nor scalable for large groups of users. In response, we introduce PairSonic, a novel group pairing protocol that extends trust from physical encounters to online communication. PairSonic simplifies the pairing process by automating the tedious verification tasks of previous methods through an acoustic out-of-band channel using smartphones’ built-in hardware. Our protocol not only facilitates connecting users for computer-supported collaboration, but also provides a more user-friendly and scalable solution to the authentication ceremonies currently used in end-to-end encrypted messengers like Signal or WhatsApp. PairSonic is available as open-source software: https://github.com/seemoo-lab/pairsonic

      November 2024  Proceedings of the ACM on Human-Computer Interaction Article

    Sounds Good? Fast and Secure Contact Exchange in Groups

    Florentin Putz, Steffen Haesler, Matthias Hollick

    PDF BibTeX DOI: 10.1145/3686964

    Abstract
    Trustworthy digital communication requires the secure exchange of contact information, but current approaches lack usability and scalability for larger groups of users. We evaluate the usability of two secure contact exchange systems: the current state of the art, SafeSlinger, and our newly designed protocol, PairSonic, which extends trust from physical encounters to spontaneous online communication. Our lab study (N=45) demonstrates PairSonic’s superior usability, automating the tedious verification tasks from previous approaches via an acoustic out-of-band channel. Although participants significantly preferred our system, minimizing user effort surprisingly decreased the perceived security for some users, who associated security with complexity. We discuss user perceptions of the different protocol components and identify remaining usability barriers for CSCW application scenarios.

      November 2024  IEEE Access Article

    Delay-Aware Online Resource Allocation for Buffer-Aided Synchronous Federated Learning Over Wireless Networks

    Jing Liu, Jing Zheng, Jing Zhang, Lin Xiang, Derrick Wing Kwan Ng, Xiaohu Ge

    BibTeX DOI: 10.1109/ACCESS.2024.3489657

    Abstract
    Synchronous federated learning (FL) over wireless networks often suffers from the straggler effect, when the time required for training local models and uploading trained parameters varies significantly across heterogeneous wireless devices. This disparity prolongs the duration needed for model aggregation at the data center and slows down the convergence of synchronous FL, posing a significant challenge for FL over wireless networks. In this paper, we propose a novel buffer-aided FL scheme to mitigate the straggler effect. A buffer with sufficiently large storage is deployed at each wireless device to temporarily store the collected training data and adaptively outputs it during local training, according to the computational capabilities and communication data rates of the wireless devices. Consequently, all local models can be synchronously aggregated at the data center to reduce the number of rounds required for model aggregation in FL. To ensure timely information updates, a staleness function is further introduced to characterize the freshness of the data used to train local models. Additionally, the entropic value-at-risk (EVaR) of the data queues is introduced to eliminate the impact of discarded data at the buffers and improve the accuracy of trained local models. We formulate a delay-aware online stochastic optimization problem to minimize the long-term average staleness of all wireless devices for buffer-aided FL. Our problem formulation simultaneously guarantees the stability of data queues at the wireless devices and reduces the risk of data loss. By employing the Lyapunov optimization technique, we transform the problem into instantaneous deterministic optimization subproblems and further solve each subproblem online via utilizing its hidden convexity. Simulation results demonstrate that the proposed buffer-aided synchronous FL scheme can effectively improve the convergence rate of FL and, at the same time, ensure timely synchronization of heterogeneous wireless devices.

      October 2024  2024 IEEE Conference on Communications and Network Security (CNS) Conference

    A Data-Driven Evaluation of the Current Security State of Android Devices

    Ernst Leierzopf, René Mayrhofer, Michael Roland, Wolfgang Studier, Lawrence Dean, Martin Seiffert, Florentin Putz, Lucas Becker, Daniel Thomas

    BibTeX DOI: 10.1109/CNS62487.2024.10735682

    Abstract
    Android’s fast-paced development cycles and the large number of devices from different manufacturers do not allow for an easy comparison between different devices’ security and privacy postures. Manufacturers each adapt and update their respective firmware images. Furthermore, images published on OEM websites do not necessarily match those installed in the field. Relevant software aspects do not remain static after initial device release, but need to be measured on real devices that receive these updates. There are various potential sources for collecting such attributes, including webscraping, crowdsourcing, and dedicated device farms. However, raw data alone is not helpful in making meaningful decisions on device security and privacy. We make a website available to access collected data. Our implementation focuses on reproducible requests and supports filtering by OEMs, devices, device models, and attributes. To improve usability, we further propose a security score grounded on the list of attributes. Based on input from Android experts, including a focus group and eight individuals, we have created a method that derives attribute weights from the importance of attributes for mitigating threats on the Android platform. We derive weights for general use cases and suggest possible examples for more specialized weights for groups of confidentiality/privacy-sensitive users and integrity-sensitive users. Since there is no one-size-fits-all setting for Android devices, our website provides the possibility to adapt all parameters of the calculated security score to individual needs.

      October 2024  5th International Conference on Resilient Systems (ICRS 2024) Conference

    Communication, Navigation and Surveillance Data Channel for UAVs using LoRa Technology

    Julian Zobel, Maximilian Bauer, Björn Scheuermann

    BibTeX DOI: 10.3929/ethz-b-000696625

    Abstract
    The application of Unmanned Aerial Systems (UAS) in crisis scenarios demands a robust wireless channel for Communication, Navigation, and Surveillance (CNS) with a ground- based operation base. This work explores the integration of the LoRa technology on small Unmanned Aerial Vehicles (UAVs) to address challenges related to limited cellular availability in crisis areas and impracticality of satellite communication in such small and constrained UAVs. LoRa provides high channel robustness, extended range, low energy consumption, and cost-effectiveness. This, however, comes with a trade-off against low throughput and strict duty cycle restrictions. This work aims at establishing and maintaining a persistent LoRa-based CNS channel for the commonly used MAVLINK protocol in both uplink and downlink direction. This requires the regulation, adaptation, and optimization of MAVLINK traffic with regards to the constraints of LoRa using traffic shaping, message segmentation, and data compression based on a traffic analysis of MAVLINK communication. Further consideration is taken for the simultaneous operation of multiple UAVs. At the current state, we employ commercially available low-price off-the-shelf components for the UAVs and the communication equipment. Specifically, we use Raspberry Pis with a LoRa-Hat as Gateways on the ground and CNS interface on the custom-build UAV, respectively. The current findings under lab conditions indicate the successful implementation of a LoRa channel for sending heartbeat and other telemetry messages to the MAVLINK operator in the downlink direction. However, uplink traffic generally occurs in larger bursts and larger message sizes, overloading the LoRa channel and requiring further improvements before its efficient application. In-flight tests are planned to assess the real-world performance of the LoRa channel, focusing on general applicability, Quality of Service (QoS) for the MAVLINK communication, and multi-gateway application for range extensions, emphasizing the adaptability and scalability of the proposed LoRa-based CNS channel in crisis scenarios.

      October 2024  Technische Universität Darmstadt Darmstadt Thesis

    Large-Scale Multi-Agent Reinforcement Learning via Mean Field Games

    Kai Cui

    PDF BibTeX DOI: 10.26083/tuprints-00028568

    Abstract
    In this dissertation, we discuss the mathematically rigorous multi-agent reinforcement learning frameworks of mean field games (MFG) and mean field control (MFC). Dynamical multi-agent control problems and their game-theoretic counterparts find many applications in practice, but can be difficult to scale to many agents. MFGs and MFC allow the tractable modeling of large-scale dynamical multi-agent control and game problems. In essence, the idea is to reduce interaction between infinitely many homogeneous agents to their anonymous distribution – the so-called mean field. This reduces many practical problems to considering a single representative agent and – by the law of large numbers – its probability law. In this thesis, we present various novel learning algorithms and theoretical frameworks of MFGs and MFC. We address existing algorithmic limitations, and also extend MFGs and MFC beyond their restriction to (i) weakly-interacting agents, (ii) all-knowing and rational agents, or (iii) homogeneity of agents. Lastly, some practical applications are briefly considered to demonstrate the usefulness of our developed algorithms. Firstly, we consider the competitive case of MFGs. There, we show that in the simplest case of finite MFGs, existing algorithms are strongly limited in their generality. In particular, the common assumption of contractive fixed-point operators is shown to be difficult to fulfill. We then contribute and analyze approximate learning algorithms for MFGs based on regularization, which allows for a trade-off between approximation and tractability. We then proceed to extend results to MFGs on graphs and hypergraphs, in order to increase the descriptiveness of MFGs and ameliorate the restriction of homogeneity. Lastly, we also extend towards the presence of both strongly interacting and many weakly-interacting agents, in order to obtain tractability for cases where some agents do not fall under the mean field approximation. Secondly, we investigate cooperative MFC. Initially, we consider an extension to environmental states under a simplifying assumption of static mean fields. Approximate optimality of an MFC solution is shown over any finite agent solution. More generally, we proceed to extend MFC to strongly interacting agents, similar to the MFG scenario. Our final extension considers partial observability, where decentralized agents act only upon available information. Here, a framework optimizing over Lipschitz classes of policies is introduced. We obtain policy gradient approximation guarantees for the latter two settings. The frameworks are verified theoretically by showing approximate optimality of MFC, and experimentally by demonstrating performance comparable or superior to state-of-the-art multi-agent reinforcement learning algorithms. Finally, we briefly explore some potential applications of MFGs and MFC in scenarios with large populations of agents. We survey applications in distributed computing, cyber-physical systems, autonomous mobility and routing, as well as natural and social sciences. We also take a closer look at two particular applications in UAV swarm control and edge computing. In the former, we consider the effect of collision avoidance as an additional constraint for MFC in embodied robot swarms. In the latter, we compare MFG and MFC results for a computational offloading scenario. Overall, in this thesis we investigate the suitability of methods based on MFC and MFC for large-scale tractable multi-agent reinforcement learning. We contribute novel learning methods and theoretical approximation frameworks, as well as study some applications. On the whole, we find that MFGs and MFC can successfully be applied to analyze large-scale control and games, with high generality and outperforming some state-of-the-art solutions.

      October 2024  32nd European Signal Processing Conference (EUSIPCO 2024) Conference

    Spatial Inference Network: Indoor Proximity Detection via Multiple Hypothesis Testing

    Martin Gölz, Luca O. Baudenbacher, Abdelhak M. Zoubir, Visa Koivunen

    PDF BibTeX

    Abstract
    Spatial inference is an important task in large-scale wireless sensor networks, the Internet of Things, radio spectrum monitoring, and smart cities. In this paper, we extend and adopt our spatial multiple hypothesis testing approach with false discovery rate control to a real-world spatial inference sensor system detecting the presence of people in indoor settings. The developed inference method is data driven, using empirical statistics and conformal p-values instead of assuming specific probability models. The approach has both, low computational complexity and energy efficient communication, hence expanding the lifespan of the network. Each sensor computes local p-values and communicates them to a fusion center. This performs the actual testing and identifies the regions where the alternative hypotheses are in place. The reliable performance of the method is demonstrated using real-world measured data acquired by an indoor wireless sensor network.

      October 2024  IEEE Access Article

    Resilience-by-Design in 6G Networks: Literature Review and Novel Enabling Concepts

    Ladan Khaloopour, Yanpeng Su, Florian Raskob, Tobias Meuser, Roland Bless, Leon Janzen, Kamyar Abedi, Marko Andjelkovic, Hekma Chaari, Pousali Chakraborty, Michael Kreutzer, Matthias Hollick, Thorsten Strufe, Norman Franchi, Vahid Jamali

    PDF BibTeX DOI: 10.1109/ACCESS.2024.3480275

    Abstract
    The sixth generation (6G) mobile communication networks are expected to intelligently integrate into various aspects of modern digital society, including smart cities, homes, health-care, transportation, and factories. While offering a multitude of services, it is likely that societies become increasingly reliant on 6G infrastructure. Any disruption to these digital services, whether due to human or technical failures, natural disasters, or terrorism, would significantly impact citizens’ daily lives. Hence, 6G networks need not only to provide high-performance services but also to be resilient in maintaining essential services in the face of potentially unknown challenges. This paper provides a general review of the state of the art on resilient systems, definitions, concepts, and approaches. Moreover, it introduces a comprehensive concept, i.e., resilience-by-design (RBD), in three different levels for designing resilient 6G communication networks, summarizing our initial studies within the German Open6GHub project. First, we outline the general RBD enabling principles and discuss their related sub-categories. Next, adopting an interdisciplinary approach, we propose to embed these principles across all 6G layers/perspectives including electronics, physical channel, network components and functions, networks, services, and cross-layer and cross-infrastructure considerations and discuss their challenges. We further elaborate the RBD principles and their realizations along with several 6G use-cases. The paper is concluded by presenting a comprehensive list of open problems for future research on 6G resilience.

      October 2024  2024 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe) Conference

    Automated Generation of Urban Medium-voltage Grids using OpenStreetMap Data

    Tobias Gebhard, Andrea Tundis, Florian Steinke

    PDF BibTeX DOI: 10.1109/ISGTEUROPE62998.2024.10863461

    Abstract
    Realistic geo-referenced electrical distribution grid (DG) models are of great importance for power system analysis and resilience studies. However, DG data are usually not publicly available. In this study, we develop a new process for the automated generation of medium-voltage (MV) grid topologies, specifically for urban areas, based on openly available data and open-source software. OpenStreetMap (OSM) data on power infrastructure, street layouts, and land use, are used as the only input source. In contrast to previous works on DG reconstruction, we use available OSM data on substation locations. Different existing methods are combined in a new, hybrid approach by considering the incompleteness of OSM data, taking the street network into account, and applying the Capacitated vehicle routing problem (CVRP) to find cost-optimal routes for power lines. Our method is tested with a German city as a case study. Furthermore, we verify the result using land use data and evaluate the quality of power-related OSM data. The results demonstrate that our approach can yield realistic geo-referenced MV grid topologies, even with incomplete OSM power data.

      October 2024  IEEE Transactions on Wireless Communications Article

    Massive MIMO Multicasting with Finite Block-length

    Xuzhong Zhang, Lin Xiang, Jiaheng Wang, Xiqi Gao

    BibTeX DOI: 10.1109/TWC.2024.3423310

    Abstract
    Massive multiple-input multiple-output (MIMO) multicasting is a promising approach for simultaneously delivering common messages to multiple users in next-generation wireless networks. However, existing studies have exclusively focused on multicast beamforming designs based on the Shannon capacity, assuming the infinite blocklength (IBL) for transmission. This assumption may lead to strictly suboptimal designs for practical multicast transmissions with finite blocklength (FBL), especially in ultra-reliable low-latency communications. In this paper, we explore the beamforming design for massive MIMO multi-group multicasting in the FBL regime. Our study considers both the max-min fairness and the weighted sum rate criteria for a comprehensive treatment. Due to the non-concave FBL rate function, the resulting optimization problems are known to be notoriously hard. We characterize the necessary and sufficient condition for the non-negative FBL rate to be a concave function of the received signal-to-interference-plus-noise ratio (SINR). Considering a finite number of transmit antennas, we propose low-complexity majorization-minimization (MM) type algorithms, which update variables in either closed or semi-closed form, to achieve locally optimal solutions of the formulated optimization problems. We further show that, as the number of transmit antennas becomes large, the optimal beamformer of each group aligns asymptotically with a linear combination of the channel vectors of that group of users, where the optimal normalized combining coefficients are derived in closed form. Subsequently, we obtain the globally optimal multicast beamformers by optimizing the power allocation using low-complexity iterative algorithms. Simulation results show that the proposed schemes outperform several existing methods, especially those employing the Shannon capacity as the performance metric. Moreover, the proposed algorithms exhibit complexities that only slightly grow with the number of transmit antennas and they can notably reduce the computation time by up to two orders of magnitude over the benchmarks, making them highly beneficial for massive MIMO applications.

      September 2024  2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW 2024) Conference

    AR-CP: Uncertainty-Aware Perception in Adverse Conditions with Conformal Prediction and Augmented Reality For Assisted Driving

    Achref Doula, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1109/CVPRW63382.2024.00026

    Abstract
    Deep learning models are pivotal in enhancing driver assistance systems and improving environmental perception. However, the tendency of neural networks towards overconfident predictions poses a risk of inaccurate predictions, potentially compromising driver safety in adverse conditions. To mitigate this issue, we introduce AR-CP, an uncertainty-aware framework designed to augment driver perception in scenarios characterized by adverse weather and insufficient lighting, through the integration of conformal prediction and augmented reality (AR). Our framework initiates with a conformal prediction step that produces an uncertainty-aware prediction set including potential object classes at a predefined probability level. Subsequently, AR is used to provide a simplified and informative visualization of the closest common parent class of the classes in the prediction set, thereby reducing the likelihood of misinformation. We provide a principled formulation and theoretical analysis of our framework. We evaluate AR-CP on the ROAD dataset, a large dataset containing different difficult situations that induce high uncertainty during prediction time. The results show that our framework outperforms state-of-the-art approaches in providing smaller prediction sets while holding the theoretical guarantees, ensuring an uncertainty-aware prediction, and reducing user confusion. We conduct an immersive user study with 15 participants to investigate the effects of our concept on the quality of perception, situation awareness, and mental load of participants. The results show that our concept facilitates a safer driving experience while holding the mental load low and the situation awareness high.

      September 2024  IEEE Transactions on Wireless Communications Article

    User Tracking and Direction Estimation Codebook Design for IRS-Assisted mmWave Communication

    Moritz Garkisch, Sebastian Lotter, Gui Zhou, Vahid Jamali, Robert Schober

    BibTeX DOI: 10.1109/TWC.2024.3463211

    Abstract
    Future communication systems are envisioned to employ intelligent reflecting surfaces (IRSs) and the millimeter wave (mmWave) frequency band to provide reliable high-rate services. For mobile users, the time-varying channel state information (CSI) requires adequate adjustment of the reflection pattern of the IRS. We propose a novel codebook-based user tracking (UT) algorithm for IRS-assisted mmWave communication, allowing suitable reconfiguration of the IRS unit cell phase shifts, resulting in a high reflection gain. The presented algorithm acquires the direction information of the user based on a peak maximum likelihood (ML)-based direction estimation. Using the direction information, the user’s trajectory is extrapolated to proactively update the adopted codeword and adjust the IRS phase shift configuration accordingly. Furthermore, we conduct a theoretical analysis of the direction estimation error and utilize the obtained insights to design a codebook specifically optimized for direction estimation. Our results show that the proposed ML-based direction estimation algorithm outperforms a multiple signal classification (MUSIC)-based reference scheme. The proposed direction estimation codebook improves the direction estimation error for both these schemes as compared to when a reference codebook is used. Also, the accuracy of the proposed UT algorithm is shown to be competitive with Kalman filter-based UT, while the proposed scheme requires fewer a priori assumptions on the user movement. Furthermore, the average achieved signal-to-noise ratio (SNR) as well as the average effective rate of the proposed UT algorithm are analyzed. The proposed UT algorithm requires only a low overhead for direction and channel estimation and avoids outdated IRS phase shifts. Furthermore, it is shown to outperform three benchmark schemes based on direct phase shift optimization, optimal codeword selection, and hierarchical codebook search, respectively, via computer simulations.

      September 2024  38th European Conference on Object-Oriented Programming and 33rd International Symposium on Software Testing and Analysis Conference

    Library Based Choreographies in Lean

    Simon Daniel, David Richter, Mira Mezini

    BibTeX

    Abstract
    The field of parallel and distributed systems has been established for decades and the relevance of communication in the context of programming is still growing. With growing complexity, distributed software systems become increasingly challenging to maintain and reason about. In traditional programming languages, application developers must ensure that communication between nodes is correct and free of deadlocks. Multiparty and choreographic languages, address these issues by combining the code for all actors in a single specification. This approach facilitates reasoning about distributed behaviour and promises inherent guarantees of deadlock freedom. Functional programming is a paradigm that clearly separates pure mathematical functions from side-effectful code. Recently, a practical implementation of a functional choreographic programming library has been published in Haskell. This paper builds upon the choreographic concept in the functional programming language Lean by adding features and increasing the safety of distributed value access.

      September 2024  2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Conference

    Adapters Strike Back

    Jan-Martin O. Steitz, Stefan Roth

    BibTeX DOI: 10.1109/CVPR52733.2024.02213

    Abstract
    Adapters provide an efficient and lightweight mechanism for adapting trained transformer models to a variety of dif-ferent tasks. However, they have often been found to be outperformed by other adaptation mechanisms including low-rank adaptation. In this paper, we provide an in-depth study of adapters, their internal structure, as well as vari-ous implementation choices. We uncover pitfalls for using adapters and suggest a concrete, improved adapter architecture, called Adapter+, that not only outperforms previous adapter implementations but surpasses a number of other, more complex adaptation mechanisms in several challenging settings. Despite this, our suggested adapter is highly robust and, unlike previous work, requires little to no manual inter-vention when addressing a novel scenario. Adapter+ reaches state-of-the-art average accuracy on the VTAB benchmark, even without a per-task hyperparameter optimization. ††Code is available at https://github.com/visinf/adapter_plus.

      September 2024  35th International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC’24) Conference

    Outage Probability Analysis of Multi-Connectivityin UAV-Assisted Urban mmWave Communication

    Zhengxin Cao, Jing Zhang, Lin Xiang, Xiaohu Ge

    BibTeX

    Abstract
    Unmanned aerial vehicle (UAV)-assisted millimeterwave (mmWave) communication presents a promising solution for high data-rate wireless applications in urban environments. However, due to limited energy supply and communication capability, UAVs can only provide temporary communication services. This challenge motivates the exploration of three-dimensional (3D) integrated aerial and ground mmWave communications utilizing the multi-connectivity (MC) technique. By leveraging both ground and aerial mmWave links over licensed and unlicensed mmWave spectrums, respectively, the MC technique can effectively exploit spatial and frequency diversity to enhance the connectivity and reliability of UAV-assisted mmWave communications. We develop a unified framework based on stochastic geometry and Markov chain to analyze the coverage and outage probabilities of the 3D integrated aerial/ground mmWave networks. Furthermore, we show that an optimal UAV flight altitude for maximizing the coverage probability of UAV communication exists and derive it in a closed-form expression. Simulation results demonstrate that UAVs can maintain reliable mmWave connections even when connections from terrestrial mmWave base stations (BSs) are obstructed by buildings, underscoring the benefits of MC in enhancing the robustness of 3D integrated aerial and ground mmWave networks.

      September 2024  Computer Communications Article

    Progressive updates of convolutional neural networks for enhanced reliability in small satellite applications

    Olga Kondrateva, Stefan Dietzel, Maximilian Schambach, Johannes Otterbach, Björn Scheuermann

    PDF BibTeX DOI: 10.1016/j.comcom.2024.07.012

    Abstract
    Small satellites enable many important applications for both economic and scientific purposes. Many of these applications are inherently data-centric and rely on large amounts of high-resolution satellite imagery to be delivered in a timely manner. However, communicating this data to Earth is challenging due to intermittent connectivity, high packet losses, low data rates, and similar issues. Therefore, efficient onboard prioritization and data processing are essential for future satellite missions. Machine learning methods, such as deep neural networks, are very suitable for such prioritization, as they are already used extensively for satellite imagery processing and they can be deployed onboard of satellites. However, updating them to support new classification requirements when the satellite is already in orbit is difficult, as often multiple passes are required to complete model transmission due to the communication challenges. To cope with this issue, we propose a progressive transmission mechanism for model updates, which leverages vector quantization and arithmetic coding. Our mechanism allows to achieve high accuracies even with partially updated models. Evaluation results show that our mechanism significantly outperforms other less optimized transmission schemes.

      September 2024  5th International Conference on Resilient Systems (ICRS 2024) Conference

    Enhancing Short-Term Discharge Predictions: An innovative ARIMA-iGARCH Model for Improved Flood Forecasting and Disaster Resilience

    Mahshid Khazaeiathar, Britta Schmalz

    BibTeX DOI: 10.3929/ethz-b-000696625

    Abstract
    Choosing the most suitable model for discharge simulation is challenging, especially with short-term data. While artificial neural networks excel at detecting river flow patterns, they require substantial data for training, making them less effective with limited datasets. As an alternative, Autoregressive Integrated Moving Average (ARIMA) models can be utilized for short-term data. However, severe volatilities and inherent non-stationarity in hydrological time series can introduce significant errors. This study introduces a new adaptive hybrid model, ARIMA-iGARCH (Integrated Generalized AutoRegressive Conditional Heteroscedasticity), designed to account for volatility and non-stationarity, thus minimizing errors in short-term time series modeling. The ARIMA-iGARCH model specifically addresses the inconsistency of variance and non-stationary behavior in discharge time series. We applied the ARIMA-iGARCH model to four hourly discharge time series of the Schwarzbach River upstream of the gauge Nauheim in Hesse, Germany. In this process, the iGARCH model was used for prediction, and hybrid model parameters were obtained by combining ARIMA and GARCH models, assuming a normal distribution for residuals. The results demonstrate that the new adaptive hybrid model, based on this special parameter estimation method, offers less complexity, greater accuracy, and more reliable predictions. By capturing fluctuations in time series variance, the ARIMA-iGARCH model significantly improves the modeling of long-memory, non-linear, non-stationary, and particularly short-term datasets. This improvement is crucial for disaster resilience, as accurate discharge predictions enhance flood forecasting and management. Effective flood forecasting leads to better preparedness and response strategies, mitigating the impacts of hydrological disasters. In conclusion, the ARIMA-iGARCH model represents a significant advancement for hydrological time series modeling, particularly with short-term data, contributing to disaster resilience by enabling more accurate and reliable flood predictions.

      August 2024  59th IEEE Internartional Conference on Communications (ICC’24) Conference

    Joint Communication and Computing Optimization for Digital Twin Synchronization with Aerial Relay

    Yi Wang, Markus Krantzik, Lin Xiang, Anja Klein

    BibTeX DOI: 10.1109/ICC51166.2024.10622630

    Abstract
    Digital twin (DT) applications usually need to be synchronized in real time with the state of the physical system (PS). This process includes both synchronizing the data collected by sensors in the PS to a server and updating the DT at the server in adaptation to the dynamics of the PS. However, communicating with power- and rate-limited sensors and processing high-volume sensed data with limited computing resources present significant challenges for DT synchronization. In this paper, we tackle both bottlenecks by proposing a joint communication and computing design for DT synchronization. In particular, we exploit buffering at an aerial cluster head of the sensors and enable buffer-aided (BA) relaying to increase the communication throughput of the sensors during data synchronization. Moreover, we adopt a novel stream computing scheme, which allows for DT updating in parallel with data synchronization, to accelerate DT synchronization. To maximize the performance of the proposed approach, we jointly optimize the trajectory of the aerial relay and the communication and computing resource allocation for minimizing the DT synchronization time. The formulated problem is a mixed-integer nonconvex problem, which is generally intractable. To tackle this challenge, we propose a low-complexity two-layer iterative suboptimal algorithm. Our simulation results show that the DT synchronization time can be significantly reduced by up to 43.8%, through stream computing and the joint optimization of the relay’s trajectory and the communication/computing resource allocation.

      August 2024  59th IEEE Internartional Conference on Communications (ICC’24) Conference

    Age of Information Minimization in Status Update Systems with Imperfect Feedback Channel

    Friedrich Pyttel, Wanja De Sombre, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX DOI: 10.1109/ICC51166.2024.10622227

    Abstract
    Status Update System (SUS) are monitoring applications of Internet of Things (IoT). They are formed by a sender that monitors a remote process and sends status updates to a receiver over a wireless channel. For successful monitoring, the sender must keep the status updates at the receiver fresh. This freshness is generally measured using the Age of Information (AoI) metric. The aim of the sender is to find a monitoring and transmission strategy that minimizes the AoI. To find the optimal strategy, the sender needs to accurately track the AoI at the receiver, i.e., it needs to perfectly know whether a transmitted status update is correctly received or not. This knowledge can be achieved by using a feedback channel between receiver and sender to send acknowledge (ACK) or negative acknowledge (NACK) messages. However, in real applications, the feedback channel is not perfect, and the transmission of ACK/NACK messages might fail. This means, the monitoring and transmission decisions have to be made under uncertainty about the receiver’s AoI. To overcome this challenge, we introduce the concept of a socalled belief distribution and propose a joint monitoring and transmission strategy at the sender based on reinforcement learning. Our approach, termed Belief Learning, exploits the belief distribution to minimize the AoI at the receiver. Through numerical simulations we show that Belief Learning enables the sender to achieve near-optimal performance with respect to the perfect feedback channel case.

      August 2024  59th IEEE Internartional Conference on Communications (ICC’24) Conference

    Two-Sided Learning: A Techno-Economic View of Mobile Crowdsensing under Incomplete Information

    Sumedh Jitendra Dongare, Bernd Simon, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX DOI: 10.1109/ICC51166.2024.10623047

    Abstract
    In Mobile Crowdsensing (MCS) a mobile crowd-sensing platform (MCSP) collects sensing data from mobile units (MUs) in exchange for payment. The MCSP broadcasts a list of available sensing tasks. Based on this list, each MU solves a task proposal problem to decide which task it is willing to perform and sends a proposal to the MCSP. Based on the MUs’ proposals, the MCSP solves a task assignment problem. There are two challenges when finding efficient task proposal strategies for the MUs and an efficient task assignment strategy for the MCSP (i) The techno-economic perspective of MCS: From the technical perspective, MCS should maximize the data quality while minimizing time and energy consumption. From the economic perspective, there are two sides, the MUs and the MCSP which act as selfish decision-makers, who aim at maximizing their own income. (ii) Incomplete information at two sides: Initially, the MCSP does not know the expected data quality and the MUs do not know the expected effort required for task completion. To overcome these challenges, we propose a novel Two-Sided Learning (TSL) approach. At the MU side, TSL is based on an innovative gradient-based multi-armed bandit solution to maximize the MUs’ utility under incomplete information about the strategies of other MUs. At the MCSP side, a learning strategy is used to find the task assignment strategy that maximizes its utility. Simulation results show that TSL achieves near-optimal social welfare, which is the sum of MUs’ and MCSP’s utilities, and a near-optimal energy efficiency.

      August 2024  59th IEEE Internartional Conference on Communications (ICC’24) Conference

    Minimizing the Age of Incorrect Information for Status Update Systems with Energy Harvesting

    Sumedh Jitendra Dongare, Aleksandar Jovovich, Wanja De Sombre, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX DOI: 10.1109/ICC51166.2024.10622719

    Abstract
    Status Update Systems (SUSs) are central components in applications like environmental sensing or smart cities. They consist of a sender monitoring a remote process and sending the sensed information to a receiver. The sender aims to deliver fresh information about the monitored process’s state to allow the receiver to timely respond to the process’s changes. In SUSs, the sender is usually battery operated. Therefore, to increase the available energy we consider Energy Harvesting (EH). Moreover, as at the receiver the information transmitted by the sender is only relevant when the process’s state changes, we measure the information’s freshness using Age of Incorrect Information (AoII). Finding the optimal transmission strategy at the sender that minimizes the AoII requires perfect system knowledge, i.e., the behavior of the monitored process, the channel quality, and the available energy. However, in real applications this knowledge is usually not available. To overcome this challenge, we first establish the optimality of threshold-based policies for AoII minimization in SUSs with EH capabilities by proving that there exists an AoII value depending on the observed state of the monitored process, the battery level and the receiver’s estimation of the monitored process’s state beyond which transmitting is preferable over idling. Next, we exploit the threshold-based policies’ structure and deploy a learning algorithm based on Finite-Difference Policy Gradient (FDPG). Our proposed approach finds the AoII thresholds without requiring perfect system knowledge. Simulations show that our approach outperforms reference algorithms by at least 20% and efficiently learns near-optimal policies for AoII minimization.

      August 2024  59th IEEE Internartional Conference on Communications (ICC’24) Conference

    Joint Optimization of Beamforming and 3D Array-Steering for UAV-Aided ISAC

    Fengcheng Pei, Lin Xiang, Anja Klein

    BibTeX DOI: 10.1109/ICC51166.2024.10622806

    Abstract
    In this paper, we investigate unmanned aerial vehicle (UAV)-aided integrated sensing and communication (ISAC) by employing a uniform linear array (ULA) of patch antennas onboard the UAV. The three-dimensional (3D) directional gain pattern of the patch antennas and the array beamforming are jointly exploited to facilitate efficient ISAC signal transmission for sensing multiple targets and communicating with multiple users. Assuming the positions of the targets and users are known, we jointly optimize the beamforming and 3D array-steering of the patch antenna array to maximize the sum of transmit beam-pattern gains towards the targets while guaranteeing quality-of-service (QoS) for each communication user. The formulated optimization problem is nonconvex and generally intractable. Exploiting the special structures underlying the problem, we propose a low-complexity iterative algorithm based on proximal block coordinate descent (BCD) to decompose the problem into several convex and manifold optimization subproblems and iteratively solve them. Simulation results verify the benefits of joint beamforming and 3D steering optimization for UAV-aided ISAC using patch antenna array, particularly when the communication QoS requirements are stringent.

      August 2024  The 12th IFAC Symposium on Control of Power & Energy Systems Conference

    Robust and Chance-Constrained Dispatch Policies for Linear Power Systems

    Hans Stenglein, Timm Faulwasser, Florian Steinke

    PDF BibTeX DOI: 10.1016/j.ifacol.2024.07.463

    Abstract
    Several methods have recently been proposed to determine optimal feedback control policies for generator dispatch under uncertain grid in-feeds. This work presents and explains two such approaches based on a constrained linear power flow model in a unified notation, a robust method based on linear programming and a chance-constrained one based on polynomial chaos expansions and second order cone programming. This work compares both methods on the same small test example and discusses consequences for long-term grid expansion planning. While the robust method is easier to implement, it is also not more restrictive. For the chance-constrained method, we highlight a serious drawback in reformulations common in the literature. As an aside, we extend the existing formulation to cases where not all disturbances can be measured, and combine it with a robust pre-processing scheme.

      August 2024  59th IEEE Internartional Conference on Communications (ICC’24) Conference

    Joint Beamforming and Trajectory Optimization for UAV-Enabled ISAC under a Finite Energy Budget

    Burak Yilmaz, Lin Xiang, Anja Klein

    BibTeX DOI: 10.1109/ICCWorkshops59551.2024.10615353

    Abstract
    This paper studies joint beamforming and trajectory optimization for integrated sensing and communication (ISAC) enabled by an energy-constrained unmanned aerial vehicle (DAV). The DAV transmits information-bearing signals using an onboard uniform linear array (ULA), for simultaneously serving downlink communication users and sensing multiple targets during its flight. To explore the synergy between UAV and ISAC, we jointly optimize the DAV’s flight trajectory, ISAC beamforming, and mission completion time for maximizing the accumulated sensing energy for ground targets under a finite energy budget for the UAV, while guaranteeing quality-of-service for communication users. The formulated problem is highly non-convex, which is generally intractable. Motivated by the success of approximate dynamic programming (DP) methods, we propose a novel low-complexity high-quality solution by combining the one-step lookahead rollout algorithm in approximate DP and the semidefinite programming technique in convex optimization. Our simulation results show that, compared to two baseline schemes, the proposed scheme can significantly enlarge the achievable sensing and communication performance region for ISAC.

      July 2024  19th International Conference on Availability, Reliability and Security (ARES2024) Conference

    Navigating the landscape of IoT security and associated risks in critical infrastructures

    Andrej Pastorek, Andrea Tundis

    BibTeX DOI: 10.1145/3664476.3669979

    Abstract
    The Internet of Things (IoT) presents transformative opportunities for connectivity and automation across various sectors, but it also introduces significant security risks that need to be comprehensively addressed. Indeed, the growing integration of IoT devices, including their vulnerabilities, into critical infrastructures amplifies potential risks in daily life, making these systems prime targets for cybercriminal activities, including espionage and sabotage. Cases where IoT devices have been misused, due to firmware vulnerabilities, embedded passwords, and hidden backdoors are real-world scenarios, that pose significant threats to privacy and security. That’s why this paper aims to point out the urgency of addressing these issues as IoT applications continue to proliferate across healthcare, transportation, urban development and other sectors. Different types of vulnerabilities and their implications with focus on urban critical infrastructures, which can lead to severe consequences like energy blackouts, water contamination, and widespread service disruptions, especially in densely populated areas, are discussed. Moreover, the need of a multidimensional approach that encompasses technological, legal, social, and economic considerations, to deal with those broader cybersecurity and risk management implications of IoT is highlighted. As a consequence, the need for continuous evolution in security strategies to keep pace with the rapid advancements in IoT technologies is pointed out, thus arguing for a proactive approach to safeguard IoT systems against emerging threats and to ensure the safe and resilient operation of these increasingly integral parts of modern critical infrastructures.

      July 2024  22nd European Control Conference Conference

    Geometric Data-Driven Dimensionality Reduction in MPC with Guarantees

    Roland Schurig, Andreas Himmel, Rolf Findeisen

    BibTeX DOI: 10.23919/ECC64448.2024.10591254

    Abstract
    We address the challenge of dimension reduction in the discrete-time optimal control problem which is solved repeatedly online within the framework of model predictive control. Our study demonstrates that a reduced-order approach, aimed at identifying a suboptimal solution within a low-dimensional subspace, retains the stability and recursive feasibility characteristics of the original problem. We present a necessary and sufficient condition for ensuring initial feasibility, which is seamlessly integrated into the subspace design process. Additionally, we employ techniques from optimization on Riemannian manifolds to develop a subspace that efficiently represents a collection of pre-specified high-dimensional data points, all while adhering to the initial admissibility constraint.

      July 2024  25th IEEE Wireless Communications and Networking Conference (WCNC2024) Conference

    Joint Optimization of Beamforming and 3D Array Steering for Multi-antenna UAV Communications

    Lin Xiang, Fengcheng Pei, Anja Klein

    BibTeX DOI: 10.1109/WCNC57260.2024.10570943

    Abstract
    In this paper, we consider unmanned aerial vehicle (UAV)-aided downlink communication using a rotatable array of directional antennas such as half-wavelength dipoles. The antenna array is mounted onboard the UAV using a gimbal and can be flexibly rotated in the three-dimensional (3D) space. As such, the directional gain pattern of dipoles and the array beamforming can be both best exploited to facilitate efficient information transmission to multiple low-priority (or secondary) users while mitigating co-channel interference for another high-priority (or primary) user coexisting with but not served by the UAV. Assuming that the direction of the high-priority user is known, we jointly optimize the electrical beamforming and mechanical steering of the rotatable dipole array for maximizing the weighted sum-rate achievable by the low-priority users while minimizing the interference power radiated in the direction of the high-priority user. The formulated optimization problem is nonconvex and generally intractable. Exploiting its special problem structure, we decompose the problem into several convex and manifold optimization subproblems and further propose a low-complexity iterative algorithm based on proximal block coordinate descent for solution. Our simulation results verify the convergence of the proposed algorithm. Moreover, compared with systems employing non-rotatable and rotatable arrays of isotropic antennas, the rotatable dipole array can flexibly adjust the gain patterns in different azimuth and elevation angles to increase the communication throughput by up to 300% and 77%, respectively.

      June 2024 Other

    Resilienz in der digitalisierten Landwirtschaft: Abhängigkeiten deutscher landwirtschaftlicher Betriebe von Kommunikations- und Energieinfrastruktur im Katastrophenschutz

    Franz Kuntke, Christian Reuter

    PDF BibTeX DOI: 10.5281/zenodo.12209183

    Abstract
    Die Landwirtschaft erfährt eine kontinuierliche Digitalisierung, wobei die Bedeutung von Daten für die eingesetzten Werkzeuge zunimmt. Im Gegensatz zu anderen kritischen Infrastrukturen hat der durchschnittliche landwirtschaftliche Betrieb eine geringe Anzahl von Mitarbeiter:innen. Die Anforderungen an die Landtechnik, ihre Umsetzung und die Vorschriften unterscheiden sich deshalb von anderer KRITIS. Unklar bleiben dabei die Auswirkungen aktueller Trends wie Smart Farming auf die Widerstandsfähigkeit des Sektors und Abhängigkeiten von anderen Infrastrukturen. Einige Aspekte der landwirtschaftlichen Digitalisierung müssen dabei kritisch betrachtet werden, um hohe Sicherheitsrisiken in Zukunft zu vermeiden: Produkte müssen sichere Voreinstellungen haben und auch die Notwendigkeit von Cloud-Anbindung sollte häufiger hinterfragt werden – sowohl für eine stärkere Sicherheit als auch Resilienz gegenüber Infrastrukturausfällen und dem hohen Datenschutzbedürfnis in der Landwirtschaft. Mit richtigen Entwicklungen kann dabei die Digitalisierung nicht nur sicher gestaltet werden, sondern auch positiv auf die Resilienz und Effizienz der Betriebe wirken.

      June 2024  22nd Annual International Conference on Mobile Systems, Applications and Services (MOBISYS ‘24) Conference

    Always On Air: Adaptive Physical Layer Switching For Uninterrupted UAV Air-to-Ground Communication

    Vincenz Mechler, Felix Wiegand, Matthias Hollick, Bastian Bloessl

    BibTeX DOI: 10.1145/3661810.3663467

    Abstract
    Reliable wireless communication is crucial for remote operation of Unmanned Aerial Vehicles (UAVs). Yet, staying in control of the vehicle at all times poses a great challenge, given the dynamics of the wireless channel. Existing technologies are optimized for a given application and, therefore, not well suited for this use case, as they cannot provide both high throughput in high Signal to Noise Ratio (SNR) regimes and high reliability in low SNR regimes. To overcome this limitation, we propose a channel-aware predictive physical layer switching algorithm, utilizing the UAV’s telemetry data for implicit synchronization. We evaluate our system experimentally in a fully emulated testbed, achieving an overall outage probability as low as 0.7 % while increasing the average throughput.

      June 2024  22nd Annual International Conference on Mobile Systems, Applications and Services Conference

    Leveraging Apple’s Find My Network for Large-Scale Distributed Sensing

    Max Granzow, Alexander Heinrich, Matthias Hollick, Marco Zimmerling

    PDF BibTeX DOI: 10.1145/3643832.3661412

    Abstract
    Find My is a crowd-sourced network of hundreds of millions of Apple devices that use Bluetooth Low Energy (BLE) to detect and track the location of items. We explore the limits and opportunities of using this proprietary network for large-scale distributed sensing. The key idea is to let low-cost sensing devices emit specially crafted BLE advertisements that trick nearby Apple devices into generating location reports that carry arbitrary sensor data, which can then be retrieved from the Apple servers. This paper reports on our ongoing work to reverse engineer the Find My system and to design a protocol for the efficient and reliable collection of data from sensing devices via the Find My network. Preliminary results from real-world experiments demonstrate the feasibility of our approach and a several-fold performance improvement compared with the state of the art.

      June 2024  IEEE Transactions on Biomedical Engineering Article

    Emergency Response Person Localization and Vital Sign Estimation Using a Semi-Autonomous Robot Mounted SFCW Radar

    Christian A. Schroth, Christian Eckrich, Ibrahim Kakouche, Stefan Fabian, Oskar von Stryk, Abdelhak M. Zoubir, Michael Muma

    BibTeX DOI: 10.1109/TBME.2024.3350789

    Abstract
    The large number and scale of natural and man-made disasters have led to an urgent demand for technologies that enhance the safety and efficiency of search and rescue teams. Semi-autonomous rescue robots are beneficial, especially when searching inaccessible terrains, or dangerous environments, such as collapsed infrastructures. For search and rescue missions in degraded visual conditions or non-line of sight scenarios, radar-based approaches may contribute to acquire valuable, and otherwise unavailable information. This article presents a complete signal processing chain for radar-based multi-person detection, 2D-MUSIC localization and breathing frequency estimation. The proposed method shows promising results on a challenging emergency response dataset that we collected using a semi-autonomous robot equipped with a commercially available through-wall radar system. The dataset is composed of 62 scenarios of various difficulty levels with up to five persons captured in different postures, angles and ranges including wooden and stone obstacles that block the radar line of sight. Ground truth data for reference locations, respiration, electrocardiogram, and acceleration signals are included.

      May 2024  21st Information Systems for Crisis Response and Management Conference Conference

    Understanding Crisis Preparedness: Insights from Personal Values, Beliefs, Social Norms, and Personal Norms

    Felix Burger, Jasmin Haunschild, Christian Reuter

    PDF BibTeX DOI: 10.59297/m9nzxn14

    Abstract
    In disasters and crises, crisis preparation at the household level is crucial for societal resilience. However, motivating citizens to take personal responsibility and to implement preparedness measures remains challenging. This study investigates the impact of values, beliefs and norms on preparedness. Through a survey in Germany on resource-oriented crisis preparedness and participants’ values, a group of more prepared and less prepared individuals was differentiated. We quantitatively analyzed the correlation between values and crisis preparedness. A quantitative analysis revealed a significant association between the value of benevolence and higher preparedness. Follow-up interviews (N=30) explored participants' beliefs and norms, showing minimal group differences. Although crisis preparedness is generally viewed positively, strong negativity exists towards perceived “hoarders”. The study discusses implication for crisis communication and the design of nudges and persuasive systems, emphasizing intervention that can trigger benevolence to foster preparedness and appeal to the social benefits to counter the narrative of anti-social crisis preparedness.

      May 2024  IEEE Open Journal of Signal Processing Article

    Robust Regularized Locality Preserving Indexing for Fiedler Vector Estimation

    Aylin Tastan, Michael Muma, Abdelhak M. Zoubir

    BibTeX DOI: 10.1109/OJSP.2024.3400683

    Abstract
    The Fiedler vector is the eigenvector associated with the algebraic connectivity of the graph Laplacian. It is central to graph analysis as it provides substantial information to learn the latent structure of a graph. In real-world applications, however, the data may be subject to heavy-tailed noise and outliers which deteriorate the structure of the Fiedler vector estimate and lead to a breakdown of popular methods. Thus, we propose a Robust Regularized Locality Preserving Indexing (RRLPI) Fiedler vector estimation method that approximates the nonlinear manifold structure of the Laplace Beltrami operator while minimizing the impact of outliers. To achieve this aim, an analysis of the effects of two fundamental outlier types on the eigen-decomposition of block affinity matrices is conducted. Then, an error model is formulated based on which the RRLPI method is developed. It includes an unsupervised regularization parameter selection algorithm that leverages the geometric structure of the projection space. The performance is benchmarked against existing methods in terms of detection probability, partitioning quality, image segmentation capability, robustness and computation time using a large variety of synthetic and real data experiments.

      May 2024  12th International Conference on Learning Representations (ICLR 2024) Conference

    Learning Mean Field Games on Sparse Graphs: A Hybrid Graphex Approach

    Christian Fabian, Kai Cui, Heinz Koeppl

    PDF BibTeX

    Abstract
    Learning the behavior of large agent populations is an important task for numerous research areas. Although the field of multi-agent reinforcement learning (MARL) has made significant progress towards solving these systems, solutions for many agents often remain computationally infeasible and lack theoretical guarantees. Mean Field Games (MFGs) address both of these issues and can be extended to Graphon MFGs (GMFGs) to include network structures between agents. Despite their merits, the real world applicability of GMFGs is limited by the fact that graphons only capture dense graphs. Since most empirically observed networks show some degree of sparsity, such as power law graphs, the GMFG framework is insufficient for capturing these network topologies. Thus, we introduce the novel concept of Graphex MFGs (GXMFGs) which builds on the graph theoretical concept of graphexes. Graphexes are the limiting objects to sparse graph sequences that also have other desirable features such as the small world property. Learning equilibria in these games is challenging due to the rich and sparse structure of the underlying graphs. To tackle these challenges, we design a new learning algorithm tailored to the GXMFG setup. This hybrid graphex learning approach leverages that the system mainly consists of a highly connected core and a sparse periphery. After defining the system and providing a theoretical analysis, we state our learning approach and demonstrate its learning capabilities on both synthetic graphs and real-world networks. This comparison shows that our GXMFG learning algorithm successfully extends MFGs to a highly relevant class of hard, realistic learning problems that are not accurately addressed by current MARL and MFG methods.

      May 2024  12th International Conference on Learning Representations (ICLR 2024) Conference

    Learning Decentralized Partially Observable Mean Field Control for Artificial Collective Behavior

    Kai Cui, Sascha Hauck, Christian Fabian, Heinz Koeppl

    PDF BibTeX

    Abstract
    Recent reinforcement learning (RL) methods have achieved success in various domains. However, multi-agent RL (MARL) remains a challenge in terms of decentralization, partial observability and scalability to many agents. Meanwhile, collective behavior requires resolution of the aforementioned challenges, and remains of importance to many state-of-the-art applications such as active matter physics, self-organizing systems, opinion dynamics, and biological or robotic swarms. Here, MARL via mean field control (MFC) offers a potential solution to scalability, but fails to consider decentralized and partially observable systems. In this paper, we enable decentralized behavior of agents under partial information by proposing novel models for decentralized partially observable MFC (Dec-POMFC), a broad class of problems with permutation-invariant agents allowing for reduction to tractable single-agent Markov decision processes (MDP) with single-agent RL solution. We provide rigorous theoretical results, including a dynamic programming principle, together with optimality guarantees for Dec-POMFC solutions applied to finite swarms of interest. Algorithmically, we propose Dec-POMFC-based policy gradient methods for MARL via centralized training and decentralized execution, together with policy gradient approximation guarantees. In addition, we improve upon state-of-the-art histogram-based MFC by kernel methods, which is of separate interest also for fully observable MFC. We evaluate numerically on representative collective behavior tasks such as adapted Kuramoto and Vicsek swarming models, being on par with state-of-the-art MARL. Overall, our framework takes a step towards RL-based engineering of artificial collective behavior via MFC.

      May 2024  i-com - Journal of Interactive Media Article

    Exploring the evolving landscape of human-centred crisis informatics: current challenges and future trends

    Marc-André Kaufhold

    BibTeX DOI: 10.1515/icom-2024-0002

    Abstract
    Modern Information and Communication Technology (ICT) has been used in safety-critical situations for over twenty years. Rooted in Human-Computer Interaction (HCI) and related disciplines, the field of crisis informatics made considerable efforts to investigate social media use and role patterns in crises, facilitate the collection, processing and refinement of social media data, design and evaluate supportive ICT, and provide cumulative and longitudinal research. This narrative review examines contemporary challenges of human-centred crisis informatics and envision trends for the following decade, including (I) a broadening scope of crisis informatics, (II) the professionalisation of cross-platform collaboration of citizen communities and emergency services, (III) expert interfaces for explainable and multimodal artificial intelligence for user-generated content assessment, (IV) internet of things and mobile apps for bidirectional communication and warnings in disruption-tolerant networks, as well as (V) digital twins and virtual reality for the effective training of multi-agency collaboration in hybrid hazards.

      April 2024  Information Systems Frontiers Article

    Digital Resilience in Dealing with Misinformation on Social Media during COVID-19: A Web Application to Assist Users in Crises

    Stefka Schmid, Katrin Hartwig, Robert Cieslinski, Christian Reuter

    PDF BibTeX DOI: 10.1007/s10796-022-10347-5

    Abstract
    In crises such as the COVID-19 pandemic, it is crucial to support users when dealing with social media content. Considering digital resilience, we propose a web app based on Social Network Analysis (SNA) to provide an overview of potentially misleading vs. non-misleading content on Twitter, which can be explored by users and enable foundational learning. The latter aims at systematically identifying thematic patterns which may be associated with misleading information. Additionally, it entails reflecting on indicators of misleading tweets which are proposed to approach classification of tweets. Paying special attention to non-expert users of social media, we conducted a two-step Think Aloud study for evaluation. While participants valued the opportunity to generate new knowledge and the diversity of the application, qualities such as equality and rapidity may be further improved. However, learning effects outweighed individual costs as all users were able to shift focus onto relevant features, such as hashtags, while readily pointing out content characteristics. Our design artifact connects to learning-oriented interventions regarding the spread of misleading information and tackles information overload by a SNA-based plug-in.

      March 2024  Proceedings of the AAAI Conference on Artificial Intelligence Conference

    Learning Discrete-Time Major-Minor Mean Field Games

    Kai Cui, Gökce Dayanikli, Mathieu Laurière, Matthieu Geist, Oliver Pietquin, Heinz Koeppl

    BibTeX DOI: 10.1609/aaai.v38i9.28818

    Abstract
    Recent techniques based on Mean Field Games (MFGs) allow the scalable analysis of multi-player games with many similar, rational agents. However, standard MFGs remain limited to homogeneous players that weakly influence each other, and cannot model major players that strongly influence other players, severely limiting the class of problems that can be handled. We propose a novel discrete time version of major-minor MFGs (M3FGs), along with a learning algorithm based on fictitious play and partitioning the probability simplex. Importantly, M3FGs generalize MFGs with common noise and can handle not only random exogeneous environment states but also major players. A key challenge is that the mean field is stochastic and not deterministic as in standard MFGs. Our theoretical investigation verifies both the M3FG model and its algorithmic solution, showing firstly the well-posedness of the M3FG model starting from a finite game of interest, and secondly convergence and approximation guarantees of the fictitious play algorithm. Then, we empirically verify the obtained theoretical results, ablating some of the theoretical assumptions made, and show successful equilibrium learning in three example problems. Overall, we establish a learning framework for a novel and broad class of tractable games.

      March 2024  12th International Conference on Learning Representations (ICLR 2024) Conference

    Common Sense Initialization of Mixture Density Networks for Motion Planning with Overestimated Number of Components.

    Thomas Kreutz, Max Mühlhäuser, Alejandro Sanchez Guinea

    PDF BibTeX

    Abstract
    Mixture density networks (MDNs) are a natural choice to model multi-modal predictions for trajectory prediction or motion planning. However, MDNs are often difficult to train due to mode collapse and a need for careful initialization, which becomes even more problematic when the number of mixture components are strongly overestimated. To address this issue in motion planning problems, we propose a pre-training scheme for MDNs called common sense initialization (CSI). Pre-training with CSI allows variety-encouraging optimization such as Winner-Takes-All (WTA) to exploit the initialized weights during training, so that the MDN can converge when the number of components are overestimated. This paper presents empirical evidence for the effectiveness of CSI when applied to motion planning of pedestrian agents in urban environments.

      March 2024  Journal of Business Venturing Article

    Flip the tweet - the two-sided coin of entrepreneurial empathy and its ambiguous influence on new product development

    Konstantin Kurz, Carolin Bock, Leonard Hanschur

    PDF BibTeX DOI: 10.1016/j.jbusvent.2023.106378

    Abstract
    Is empathy a uniformly good thing for entrepreneurs? Contrasting the hitherto predominantly positive view advocated by the extant entrepreneurship literature, we develop a novel model of entrepreneurial empathy’s mechanisms and suggest a too-much-of-a-good-thing perspective. We empirically confirm this model using a dataset of 4425 real entrepreneurs, where we find that empathy influences entrepreneurial new product development as an essential entrepreneurial activity in an inverted U-shaped pattern. We further show that empathy’s negative effects are particularly detrimental for very anxious entrepreneurs. These findings provide strong evidence for considering entrepreneurial empathy an important but highly ambiguous success factor.

      February 2024  IEEE Transactions on Communications Article

    Optical IRSs: Power Scaling Law, Optimal Deployment, and Comparison With Relays

    Hedieh Ajam, Marzieh Najafi, Vahid Jamali, Robert Schober

    BibTeX DOI: 10.1109/TCOMM.2023.3327464

    Abstract
    The line-of-sight (LOS) requirement of free-space optical (FSO) systems can be relaxed by employing optical relays or optical intelligent reflecting surfaces (IRSs). In this paper, we show that the power reflected from FSO IRSs and collected at the receiver (Rx) lens may scale quadratically or linearly with the IRS size or may saturate at a constant value. We analyze the power scaling law for optical IRSs and unveil its dependence on the wavelength, transmitter (Tx)-to-IRS and IRS-to-Rx distances, beam waist, and Rx lens size. We also consider the impact of linear, quadratic, and focusing phase shift profiles across the IRS on the power collected at the Rx lens for different IRS sizes. Our results reveal that surprisingly the powers received for the different phase shift profiles are identical, unless the IRS operates in the saturation regime. Moreover, IRSs employing the focusing (linear) phase shift profile require the largest (smallest) size to reach the saturation regime. We also compare optical IRSs in different power scaling regimes with optical relays in terms of the outage probability, diversity and coding gains, and optimal placement. Our results show that, at the expense of a higher hardware complexity, relay-assisted FSO links yield a better outage performance at high signal-to-noise-ratios (SNRs), but optical IRSs can achieve a higher performance at low SNRs. Moreover, while it is optimal to place relays equidistant from Tx and Rx, the optimal location of optical IRSs depends on the phase shift profile and the power scaling regime they operate in.

      January 2024  2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Conference

    FDR-Controlled Portfolio Optimization for Sparse Financial Index Tracking

    Jasin Machkour, Daniel P. Palomar, Michael Muma

    PDF BibTeX DOI: 10.1109/ICASSP49660.2025.10890629

    Abstract
    In high-dimensional data analysis, such as financial index tracking or biomedical applications, it is crucial to select the few relevant variables while maintaining control over the false discovery rate (FDR). In these applications, strong dependencies often exist among the variables (e.g., stock returns), which can undermine the FDR control property of existing methods like the model-X knockoff method or the T-Rex selector. To address this issue, we have expanded the T-Rex framework to accommodate overlapping groups of highly correlated variables. This is achieved by integrating a nearest neighbors penalization mechanism into the framework, which provably controls the FDR at the user-defined target level. A real-world example of sparse index tracking demonstrates the proposed method’s ability to accurately track the S&P 500 index over the past 20 years based on a small number of stocks. An open-source implementation is provided within the R package TRexSelector on CRAN.

      January 2024  IEEE Open Journal of the Communications Society Article

    Deterministic K-Identification for MC Poisson Channel With Inter-Symbol Interference

    Mohammad Javad Salariseddigh, Vahid Jamali, Uzi Pereg, Holger Boche, Christian Deppe, Robert Schober

    BibTeX DOI: 10.1109/OJCOMS.2024.3359186

    Abstract
    Various applications of molecular communications (MCs) feature an alarm-prompt behavior for which the prevalent Shannon capacity may not be the appropriate performance metric. The identification capacity as an alternative measure for such systems has been motivated and established in the literature. In this paper, we study deterministic K-identification (DKI) for the discrete-time Poisson channel (DTPC) with inter-symbol interference (ISI), where the transmitter is restricted to an average and a peak molecule release rate constraint. Such a channel serves as a model for diffusive MC systems featuring long channel impulse responses and employing molecule-counting receivers. We derive lower and upper bounds on the DKI capacity of the DTPC with ISI when the size of the target message set K and the number of ISI channel taps L may grow with the codeword length n . As a key finding, we establish that for deterministic encoding, assuming that K and L both grow sub-linearly in n , i.e., K=2κlogn and L=2llogn with κ+4l∈[0,1) , where κ∈[0,1) is the identification target rate and l∈[0,1/4) is the ISI rate, then the number of different messages that can be reliably identified scales super-exponentially in n , i.e., ∼2(nlogn)R , where R is the DKI coding rate. Moreover, since l and κ must fulfill κ+4l∈[0,1) , we show that optimizing l (or equivalently the symbol rate) leads to an effective identification rate [bits/s] that scales sub-linearly with n . This result is in contrast to the typical transmission rate [bits/s] which is independent of n .

      January 2024  32nd European Signal Processing Conference (EUSIPCO) Conference

    False Discovery Rate Control for Gaussian Graphical Models via Neighborhood Screening

    Taulant Koka, Jasin Machkour, Michael Muma

    PDF BibTeX DOI: 10.23919/EUSIPCO63174.2024.10715414

    Abstract
    Gaussian graphical models emerge in a wide range of fields. They model the statistical relationships between variables as a graph, where an edge between two variables indicates conditional dependence. Unfortunately, well-established estimators, such as the graphical lasso or neighborhood selection, are known to be susceptible to a high prevalence of false edge detections. False detections may encourage inaccurate or even incorrect scientific interpretations, with major implications in applications, such as biomedicine or healthcare. In this paper, we introduce a nodewise variable selection approach to graph learning and provably control the false discovery rate of the selected edge set at a self-estimated level. A novel fusion method of the individual neighborhoods outputs an undirected graph estimate. The proposed method is parameter-free and does not require tuning by the user. Benchmarks against competing false discovery rate controlling methods in numerical experiments considering different graph topologies show a significant gain in performance.

      January 2024  Journal of Sensor and Actuator Networks Article

    On Maximizing the Probability of Achieving Deadlines in Communication Networks

    Benjamin Becker, Christian Oberli, Tobias Meuser, Ralf Steinmetz

    BibTeX DOI: 10.3390/jsan13010009

    Abstract
    We consider the problem of meeting deadline constraints in wireless communication networks. Fulfilling deadlines depends heavily on the routing algorithm used. We study this dependence generically for a broad class of routing algorithms. For analyzing the impact of routing decisions on deadline fulfillment, we adopt a stochastic model from operations research to capture the source-to-destination delay distribution and the corresponding probability of successfully delivering data before a given deadline. Based on this model, we propose a decentralized algorithm that operates locally at each node and exchanges information solely with direct neighbors in order to determine the probabilities of achieving deadlines. A modified version of the algorithm also improves routing tables iteratively to progressively increase the deadline achievement probabilities. This modified algorithm is shown to deliver routing tables that maximize the deadline achievement probabilities for all nodes in a given network. We tested the approach by simulation and compared it with routing strategies based on established metrics, specifically the average delay, minimum hop count, and expected transmission count. Our evaluations encompass different channel quality and small-scale fading conditions, as well as various traffic load scenarios. Notably, our solution consistently outperforms the other approaches in all tested scenarios.

      January 2024  New Journal of Physics Article

    A collective perception model for neighbor selection in groups based on visual attention mechanisms

    Kang Li, Liang Li, Roderich Groß, Shiyu Zhao

    PDF BibTeX DOI: 10.1088/1367-2630/ad1b81

    Abstract
    The mechanisms by which individuals visually perceive and select others within a group to interact with are of fundamental importance to various collective motion behaviors. However, they remain poorly understood theoretically, partially due to the challenges of modeling neurological perception systems. Here, the classic zoom-lens visual attention model is introduced into collective perception, leading to an attention-based model with only a single parameter. Numerical experiments reveal that the proposed model offers a unified mechanism for collective perception, as reflected in three aspects. First, this model describes both conspecific (e.g. flocking) and heterospecific (e.g. predator-prey) collective behaviors. Second, this model unifies the well-known topological and visibility models in the context of neighbor selection in conspecific groups. Third, in the context of prey selection in heterospecific groups, the model can simultaneously replicate the well-known confusion and oddity effects. These findings demonstrate the fundamental role of visual attention underlying a diverse array of collective motion behaviors.

      January 2024  2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Conference

    Sparse PCA with False Discovery Rate Controlled Variable Selection

    Jasin Machkour, Arnaud Breloy, Michael Muma, Daniel P. Palomar, Frédéric Pascal

    PDF BibTeX DOI: 10.1109/ICASSP48485.2024.10448237

    Abstract
    Sparse principal component analysis (PCA) aims at mapping large dimensional data to a linear subspace of lower dimension. By imposing loading vectors to be sparse, it performs the double duty of dimension reduction and variable selection. Sparse PCA algorithms are usually expressed as a trade-off between explained variance and sparsity of the loading vectors (i.e., number of selected variables). As a high explained variance is not necessarily synonymous with relevant information, these methods are prone to select irrelevant variables. To overcome this issue, we propose an alternative formulation of sparse PCA driven by the false discovery rate (FDR). We then leverage the Terminating-Random Experiments (T-Rex) selector to automatically determine an FDR-controlled support of the loading vectors. A major advantage of the resulting T-Rex PCA is that no sparsity parameter tuning is required. Numerical experiments and a stock market data example demonstrate a significant performance improvement.

      2024  41th International Conference on Machine Learning (2024) Conference

    Major-Minor Mean Field Multi-Agent Reinforcement Learning

    Kai Cui, Christian Fabian, Anam Tahir, H. Koeppl

    PDF BibTeX

    Abstract
    Multi-agent reinforcement learning (MARL) remains difficult to scale to many agents. Recent MARLusingMeanFieldControl(MFC)provides a tractable and rigorous approach to otherwise difficult cooperative MARL. However, the strict MFCassumption of many independent, weaklyinteracting agents is too inflexible in practice. We generalize MFC to instead simultaneously model many similar and few complex agents– as MajorMinor MeanField Control (M3FC). Theoretically, we give approximation results for finite agent control, and verify the sufficiency of stationary policies for optimality together with a dynamic programming principle. Algorithmically, we propose Major-Minor Mean Field MARL (M3FMARL) for finite agent systems instead of the limiting system. The algorithm is shown to approximate the policy gradient of the underlying M3FC MDP. Finally, we demonstrate its capabilities experimentally in various scenarios. We observe a strong performance in comparison to state-of-the-art policy gradient MARL methods.

      2024  IEEE International Conference on Robotics and Automation (ICRA) Conference

    A Modular Aerial System Based on Homogeneous Quadrotors with Fault-Tolerant Control

    Mengguang Li, Kai Cui, H. Koeppl

    PDF BibTeX DOI: 10.1109/ICRA57147.2024.10610414

    Abstract
    The standard quadrotor is one of the most popular and widely used aerial vehicle of recent decades, offering great maneuverability with mechanical simplicity. However, the under-actuation characteristic limits its applications, especially when it comes to generating desired wrench with six degrees of freedom (DOF). Therefore, existing work often compromises between mechanical complexity and the controllable DOF of the aerial system. To take advantage of the mechanical simplicity of a standard quadrotor, we propose a modular aerial system, IdentiQuad, that combines only homogeneous quadrotor-based modules. Each IdentiQuad can be operated alone like a standard quadrotor, but at the same time allows task-specific assembly, increasing the controllable DOF of the system. Each module is interchangeable within its assembly. We also propose a general controller for different configurations of assemblies, capable of tolerating rotor failures and balancing the energy consumption of each module. The functionality and robustness of the system and its controller are validated using physics-based simulations for different assembly configurations.

      2024  21st International Conference on Informatics in Control, Automation and Robotics Conference

    Automated mission management of small unmanned aircraft systems for critical events in urban air traffic

    Robin Müller, Maximilian Bauer

    BibTeX DOI: 10.5220/0012951500003822

    Abstract
    Unmanned aerial systems (UAS) have a great potential to benefit society. This has already been shown in many use-cases. Nevertheless the true potential lies in the upscaling of operations. Therefore a high automation level and ensured safety is needed. A common approach to adress safety in aviation is a risk analysis following by the design of procedures to mitigate the risks - so called contingency procedures. This paper presents a functional framework for automated mission management including contingencies for UAS. The framework is based on behavior trees and can be integrated with popular open source flight control software like PX4 and Ardupilot. Missions can be planned in a graphical interface using building blocks or in a Ground Control Station software like QGroundControl. The planning of contingency procedures is decoupled from the mission planning and allows for high modularity. Procedures can easily be added, modified or deleted, which is very important for certification o f operation. The functionality of the framework is validated in various simulations, testing a plethora of contingencies and missions. Flight tests are currently conducted. The code needed to use the framework can be found on the website: https://robin-mueller.github.io/auto-apms-guide/.

      2024  Media, War & Conflict Article

    Smartphone resilience: ICT in Ukrainian civic response to the Russian full-scale invasion

    Kateryna Zarembo, Michèle Knodt, Jannis Kachel

    PDF BibTeX DOI: doi:10.1177/17506352241236449

    Abstract
    In modern warfare, digitalization has blurred the line where civilian ends and military begins. Embedded in the participative warfare theoretical paradigm, this article looks into how the information and communication technologies (ICT) enable civic resilience under the conditions of the foreign armed aggression. Specifically, the authors explore how smartphones and smartphone applications empowered the Ukrainian civil society in the aftermath of the Russian full-scale invasion of 2022. Based on an online survey and semi-structured interviews, the article highlights how the device and its features not only allowed civilians to adapt to living in conditions of a constant threat, but also to respond and support the defence from the rear. The authors conclude that, while the smartphone becomes an ‘online resilience hub’, acquiring many new functions like a mobile office, an online volunteer (frontline logistics and procurement) hub, an air-threat warner, a first-hand news source and so on, its security provision functions are not unconditional and may turn to the opposite, depending on the physical circumstances on the ground as well as the virtual information battlefield.

      2024  23rd International Conference on Modelling and Applied Simulation Conference

    Safeguarding Critical Infrastructures with Digital Twins and AI

    Andrea Tundis, Oscar Hernan Ramirez Agudelo

    PDF BibTeX

    Abstract
    Critical infrastructures (CIs) are complex systems that are integral parts of our daily lives. We use them to access various services related to basic needs, to obtain water, energy and information, to move from one place to another, to work remotely, and so on. It is therefore essential, but challenging, to secure and protect them. On the other hand, digital twins (DTs) and artificial intelligence (AI) represent solid approaches that are well suited to modelling and analyzing complex systems, respectively. In this context, this work can be seen as a white paper that aims first to explore the main characteristics and limitations of DTs and AI when considered in isolation, and then to discuss how their combination as an intelligent entity - which represents a paradigm shift in the protection and resilience of CIs - might be beneficial to overcome such challenges and thus be useful to enhance their protection

      December 2023  IEEE Transactions on Signal Processing Article

    Fast and Robust Sparsity-Aware Block Diagonal Representation

    A. Taştan, Michael Muma, A. M. Zoubir

    BibTeX DOI: 10.1109/TSP.2023.3343565

    Abstract
    The block diagonal structure of an affinity matrix is a commonly desired property in cluster analysis because it represents clusters of feature vectors by non-zero coefficients that are concentrated in blocks. However, recovering a block diagonal affinity matrix is challenging in real-world applications, in which the data may be subject to outliers and heavy-tailed noise that obscure the hidden cluster structure. To address this issue, we first analyze the effect of different fundamental outlier types in graph-based cluster analysis. A key idea that simplifies the analysis is to introduce a vector that represents a block diagonal matrix as a piece-wise linear function of the similarity coefficients that form the affinity matrix. We reformulate the problem as a robust piece-wise linear fitting problem and propose a Fast and Robust Sparsity-Aware Block Diagonal Representation (FRS-BDR) method, which jointly estimates cluster memberships and the number of blocks. Comprehensive experiments on a variety of real-world applications demonstrate the effectiveness of FRS-BDR in terms of clustering accuracy, robustness against corrupted features, computation time and cluster enumeration performance.

      December 2023  9th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing Conference

    Solving FDR-Controlled Sparse Regression Problems with Five Million Variables on a Laptop

    Fabian Scheidt, Jasin Machkour, Michael Muma

    BibTeX DOI: 10.1109/CAMSAP58249.2023.10403478

    Abstract
    Currently, there is an urgent demand for scalable multivariate and high-dimensional false discovery rate (FDR)-controlling variable selection methods to ensure the reproducibility of discoveries. However, among existing methods, only the recently proposed Terminating-Random Experiments (T-Rex) selector scales to problems with millions of variables, as encountered in, e.g., genomics research. The T-Rex selector is a new learning framework based on early terminated random experiments with computer-generated dummy variables. In this work, we propose the Big T-Rex, a new implementation of T-Rex that drastically reduces its Random Access Memory (RAM) consumption to enable solving FDR-controlled sparse regression problems with millions of variables on a laptop. We incorporate advanced memory-mapping techniques to work with matrices that reside on solid-state drive and two new dummy generation strategies based on permutations of a reference matrix. Our numerical experiments demonstrate a drastic reduction in memory demand and computation time. We showcase that the Big T-Rex can efficiently solve FDR-controlled Lasso-type problems with five million variables on a laptop in thirty minutes. Our work empowers researchers without access to high-performance clusters to make reproducible discoveries in large-scale high-dimensional data.

      December 2023  9th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing Conference

    The Informed Elastic Net for Fast Grouped Variable Selection and FDR Control in Genomics Research

    Jasin Machkour, Michael Muma, Daniel P. Palomar

    BibTeX DOI: 10.1109/CAMSAP58249.2023.10403489

    Abstract
    Modern genomics research relies on genome-wide association studies (GWAS) to identify the few genetic variants among potentially millions that are associated with diseases of interest. Only reproducible discoveries of groups of associations improve our understanding of complex polygenic diseases and enable the development of new drugs and personalized medicine. Thus, fast multivariate variable selection methods that have a high true positive rate (TPR) while controlling the false discovery rate (FDR) are crucial. Recently, the T-Rex+GVS selector, a version of the T-Rex selector that uses the elastic net (EN) as a base selector to perform grouped variable election, was proposed. Although it significantly increased the TPR in simulated GWAS compared to the original T-Rex, its comparably high computational cost limits scalability. Therefore, we propose the informed elastic net (IEN), a new base selector that significantly reduces computation time while retaining the grouped variable selection property. We quantify its grouping effect and derive its formulation as a Lasso-type optimization problem, which is solved efficiently within the T-Rex framework by the terminated LARS algorithm. Numerical simulations and a GWAS study demonstrate that the proposed T-Rex+GVS (IEN) exhibits the desired grouping effect, reduces computation time, and achieves the same TPR as T-Rex+GVS (EN) but with lower FDR, which makes it a promising method for large-scale GWAS.

      December 2023  2023 International Conference on Embedded Wireless Systems and Networks Conference

    Demos: Robust Orchestration for Autonomous Networking

    Andreas Biri, Marco Zimmerling, Lothar Thiele

    PDF BibTeX

    Abstract
    Research in wireless sensor networks has resulted in a re- markable breadth of highly capable systems. However, while specialized protocols perform well in the setting they were designed for, they often lack the ability to quickly adapt once operating conditions change drastically. Of particular impor- tance is resilience to node and link failures, as clusters of nodes that lost their leader or split apart need to re-organize and find each other again. With Demos, we present a low- power wireless protocol that ensures robust network orches- tration despite such failures. Demos rapidly finds consen- sus on leadership with its cluster coordination mechanism even if the set of nodes fluctuates by introducing new elec- tion quorums. In addition, a novel cluster discovery scheme enables autonomous clusters to merge on the fly and max- imize network coverage. Experiments with controlled mo- bility on a multi-hop network of 24 nodes demonstrate that Demos maintains a reliable data exchange despite severe disruptions and adapts to changes within seconds. We fur- ther find that Demos’ ability to continuously coordinate and discover achieves highly robust orchestration of fully au- tonomous clusters.

      December 2023  2023 IEEE Global Communications Conference Conference

    Federated Deep Reinforcement Learning for Task Participation in Mobile Crowdsensing

    Sumedh Dongare, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX DOI: 10.1109/GLOBECOM54140.2023.10436786

    Abstract
    Mobile Crowdsensing (MCS) is a promising distributed sensing architecture that harnesses the power of sensors on mobile units (MUs) to perform sensing tasks. The MCS is a dynamic system in which the requirements of the sensing tasks, the MUs’ conditions and the available resources change over time. The performance of an MCS system depends on the selection of the MUs participating in each sensing task. However, this is not a trivial problem. An optimal task participation strategy requires non-causal knowledge about the dynamic MCS system, a requirement that cannot be fulfilled in real implementations. Moreover, centralized optimization-based approaches do not scale with increasing number of participating MUs and often ignore the MUs’ preferences. To overcome these challenges, in this paper we propose a novel multi-agent federated deep reinforcement learning algorithm (FDRL-PPO) which does not need this perfect non-causal knowledge, but instead, enables the MUs to learn their own task participation strategies based on their own conditions, available resources, and preferences. Through federated learning, the MUs share their learned strategies without disclosing sensitive information, enabling a robust and scalable task participation scheme. Numerical evaluations validate the effectiveness and efficiency of FDRL-PPO in comparison with reference schemes.

      December 2023  2023 IEEE Global Communications Conference Conference

    A Unified Approach to Learn Transmission Strategies Using Age-Based Metrics in Point-to-Point Wireless Communication

    Wanja De Sombre, Felipe Marques, Friedrich Pyttel, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX

    Abstract
    Based on the Age of Information as an optimization criterion, proposals for further age-based metrics have been made in recent years in the Internet of Things (IoT) domain. The research community’s great interest in age-based metrics for point-to-point wireless communication has led to a multitude of different scenarios being investigated, including energy opti- mization, sensing, and risk-sensitivity. All these scenarios involve a sender-receiver pair and revolve around finding appropriate times for the sender to communicate status updates to the receiver. We propose a unified and modular framework that represents the aforementioned options in various combinations and enables transferring solutions developed for specific cases to a variety of scenarios. We generalize an existing optimization approach, which decides to transmit based on a threshold for the age-based metric, using this framework. We develop a unified and extended Q-learning-based algorithm with mechanisms to learn suitable solutions for all scenarios derived from our framework. These mechanisms accelerate the learning process and result in improved algorithmic performance compared to traditional Q-learning. Furthermore, we demonstrate the effectiveness of our solution in numerical simulations. Our unified solution outperforms several reference schemes in terms of age-based metrics, energy consumption, and risk. We present our findings as a starting point to investigate transmission strategies for more general settings with a more efficient approach.

      December 2023  22nd IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2023 ) Conference

    “Can You Handle the Truth?”: Investigating the Effects of AR-Based Visualization of the Uncertainty of Deep Learning Models on Users of Autonomous Vehicles

    Achref Doula, Lennart Schmidt, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1109/ismar59233.2023.00040

    Abstract
    The recent advances in deep learning have paved the way for autonomous vehicles (AVs) to take charge of more complex tasks in the navigation process. However, predictions of deep learning models are subject to different types of uncertainty that may put the user and the surrounding environment in danger. In this paper, we investigate the effects that AR-based visualizations of 3 types of uncertainties in deep learning modules for path planning in AVs may have on drivers. The uncertainty types of the deep learning models that we consider are: the waypoint uncertainty, the situation uncertainty, and the path uncertainty. We propose 3 concepts to visualize the 3 uncertainty types on a Windshield display. We evaluate our AR-based concepts with a user study (N=20) using a VR-based immersive environment, to ensure the security of the participants. The results of our evaluation reveal that the absence of uncertainty visualization leads to lower driver engagement. More importantly, the combination of situation uncertainty and path uncertainty visualizations leads to higher driver engagement, and higher trust in the automated vehicle, while inducing an acceptable mental load for the drive.

      December 2023  2023 IEEE Global Communications Conference Conference

    Contextual Multi-Armed Bandits for Non-Stationary Heterogeneous Mobile Edge Computing

    Maximilian Wirth, Andrea Patricia Ortiz Jimenez, Anja Klein

    PDF BibTeX DOI: 10.1109/GLOBECOM54140.2023.10437572

    Abstract
    Base station (BS) selection for task offloading in Mobile Edge Computing (MEC) is a challenging problem due to the dynamic nature of MEC systems. The wireless channel as well as the load of BSs are stochastic quantities that can change in a statistically non-stationary fashion. Moreover, the computation capabilities of the BSs are heterogeneous. As the dynamic behaviour of a MEC system is, in practical scenarios, not known in advance, deciding where to offload has to be done under uncertainty about the MEC system and considering its non-stationary and heterogeneous characteristics. This paper in- vestigates latency minimization in MEC with heterogeneous BSs. In order to meet low latency demands, a mobile unit (MU) has to quickly identify the best BS for offloading different computation tasks while facing uncertainty about the non-stationary system dynamics. To solve this problem, we propose a novel piece-wise stationary contextual Multi-Armed Bandit (MAB) algorithm that treats different task types as context and detects non-stationary changes in the BSs’ performance. With the use of extensive simulations, we show that our proposed approach outperforms state-of-the-art algorithms, as it quickly adapts to changes in the MEC system and exhibits no penalty during stationary phases.

      December 2023  2023 IEEE Global Communications Conference Conference

    Robust Dynamic Trajectory Optimization for UAV-Aided Localization of Ground Target

    Lin Xiang, Mengshuai Zhang, Anja Klein

    PDF BibTeX DOI: 10.1109/GLOBECOM54140.2023.10437337

    Abstract
    In this paper, we consider employing an unmanned aerial vehicle (UAV) equipped with an onboard radar transceiver to localize a ground target at an unknown position. Exploiting the UAV’s mobility, we aim to gather line-of-sight (LoS) range measurements from favorable waypoints and improve the ensuing multi-lateration process while estimating the target’s location. To this end, we introduce a novel localization error metric, characterized geometrically by the radius of a defined confidence region where the target resides at a predetermined confidence level. Additionally, we investigate robust dynamic optimization of the UAV’s trajectory to minimize the defined localization error metric online, utilizing sequentially available but delayed range estimates. The formulated optimization problem belongs to a convex-nonconcave minimax problem, which is generally intractable. To solve this problem, we further propose two iterative online algorithms based on semidefinite programming (SDP) relaxation and alternating/sequential convex optimization techniques. Simulation results show that the proposed online schemes outperform several benchmarks, either in the final localization accuracy or in the rate of decreasing the localization error.

      December 2023  Journal of Information Security and Applications Article

    From the detection towards a pyramidal classification of terrorist propaganda

    Andrea Tundis, Ahmed Ali Shams, Max Mühlhäuser

    PDF BibTeX DOI: 10.1016/j.jisa.2023.103646

    Abstract
    With over 3,81 billion users from across the world, social media platforms provide a borderless environment for people of different nationalities, races, ethnicities, and religious beliefs to interact and communicate with each other. Not only for legitimate purposes are these digital tools used, but also groups of extremists and terrorist organizations take advantage of the features of these platforms to spread radicalization, propaganda, brainwashing and for online recruitment. Recent studies, conducted in this area, are trying to face with this phenomenon, but due to the heterogeneity of the sources, the large amount of daily data generated, and especially the different levels of radicalization of the users, make it even more difficult to define effective and general countermeasures against such phenomenon. In this context, the paper provides a solution that not only aims to support the detection of terrorist propaganda (and related users), but also to support its further categorization, centered on a pyramid classification model, by analyzing the level of users’ radicalization. This approach has two fundamental complementary advantages, as on the one hand it enables the establishment of priorities in terms of intervention, and on the other hand to define and apply targeted countermeasures based on the level of user’s radicalization. The proposed model and the results obtained from its experimentation are shown and discussed in comparison to previous works.

      November 2023  2023 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR’23) Conference

    Requirements and challenges for autonomy and assistance functions for ground rescue robots in reconnaissance missions

    Kevin Daun, Oskar von Stryk

    BibTeX DOI: 10.1109/SSRR59696.2023.10499930

    Abstract
    While rescue robots are becoming more established as part of disaster response, they are typically teleoperated in actual disasters. (Autonomous) assistance functions can improve performance, extend functionality and reduce operator overload. It is necessary to understand relevant requirements to ensure that developed capabilities apply to real-world needs. Previous analyses focused on general aspects of rescue robots, leaving a gap in understanding requirements for (autonomous) assistance functions. We address this gap and provide a detailed, evidence-driven analysis of application requirements and research challenges for (autonomous) assistance functions for rescue robots in reconnaissance missions. We base our analysis on a comprehensive model for technology acceptance and consider reports of past deployments, related analyses, our own experience from deploying robots, and insights from workshops with first responders. We define relevant aspects of an integrated function capability and analyze general and specific requirements for assistance functions and autonomy. We relate our results with current assistance functions and identify several research challenges. A key insight is the need for an increased research focus on novel approaches combining the complementary capabilities of human operators and robotic assistance functions.

      November 2023  2023 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR’23) Conference

    Affordance-based Actionable Semantic Mapping and Planning for Mobile Rescue Robots

    Frederik Bark, Kevin Daun, Oskar von Stryk

    PDF BibTeX DOI: 10.1109/SSRR59696.2023.10499938

    Abstract
    Autonomous and tele-operation of rescue robots in urban search and rescue (USAR) environments is very challenging as details of missions and environments are usually unknown, mission goals might change dynamically and there is only little repeatability between different missions. Therefore, we propose a novel actionable semantic mapping and planning approach which leverages complementary capabilities of operator and robotic assistance functions. While related methods often focus on accuracy for geometric or semantic representations, we propose a novel framework focusing on an actionable map representation which is well suited for planning complex behaviors in uncertain environments. We represent the environment topologically as a scene graph coupled with a geometrically and semantically dense representation as Truncated Signed Distance Functions. We propose to apply the concept of affordances to map possible actions and costs to object classes. Building on that, we propose a combined topological and geometric task planning method allowing for easy operator interaction on task selection and prioritization. The successful application in two complex scenarios demonstrates the flexibility and efficiency of the proposed approach and the benefit of operator interaction.

      November 2023  2023 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR’23) Conference

    Hector UI: A Flexible Human-Robot User Interface for (Semi-)Autonomous Rescue and Inspection Robots

    Stefan Fabian, Oskar von Stryk

    PDF BibTeX DOI: 10.1109/SSRR59696.2023.10499954

    Abstract
    The remote human operator’s user interface (UI) is an important link to make the robot an efficient extension of the operator’s perception and action. In rescue applications, several studies have investigated the design of operator interfaces based on observations during major robotics competitions or field deployments. Based on this research, guidelines for good interface design were empirically identified. The investigations on the UIs of teams participating in competitions are often based on external observations during UI application, which may miss some relevant requirements for UI flexibility. In this work, we present an open-source and flexibly configurable user interface based on established guidelines and its exemplary use for wheeled, tracked, and walking robots. We explain the design decisions and cover the insights we have gained during its highly successful applications in multiple robotics competitions and evaluations. The presented UI can also be adapted for other robots with little effort and is available as open source.

      November 2023  2023 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR’23) Conference

    Online 2D-3D Radiation Mapping and Source Localization using Gaussian Processes with Mobile Ground Robots

    Jonas Süß, Martin Volz, Kevin Daun, Oskar von Stryk

    BibTeX DOI: 10.1109/SSRR59696.2023.10499940

    Abstract
    We present a novel method for online radiation mapping and source localization in 2D and 3D with mobile ground robots using Gaussian Processes to assist personnel in potentially dangerous scenarios such as nuclear catastrophes or dismantling nuclear reactors. While existing methods typically make strong model assumptions or are limited for robot onboard application by high computational cost, we propose a method that requires only weak model assumptions and gains efficiency by pre-sampling and local map update schemes. The resulting models can predict the radiation levels in complex indoor environments with multiple sources and quantify the uncertainty in their estimates. The proposed method can be applied in combination with teleoperated, semi-autonomous, or autonomous exploration. It was successfully evaluated at the EnRicH 2023 competition in a decommissioned nuclear power plant, where it provided the best localization and mapping of five radiation sources and received the award for radiation mapping. Our evaluation of data from the competition validates the accuracy and computational efficiency of the proposed approach. Moreover, we provide an open-source ROS implementation of the proposed method and open-access evaluation data.

      November 2023  33rd International Conference on Field-Programmable Logic and Applications Conference

    Optimization Techniques for Hestenes-Jacobi SVD on FPGAs

    Lukas Stasytis, Zsolt István

    BibTeX DOI: 10.1109/FPL60245.2023.00028

    Abstract
    Matrix decomposition, such as the Singular Value Decomposition (SVD) is an important compute-intensive task in a wide variety of fields, from radar and simulation to image processing and compression. In light of growing data sizes, accelerators, such as FPGAs, are often considered for SVD, with the goal of increasing the compute efficiency. However, to achieve high-performance computation of SVD, we need high parallelism. For that, a thorough reevaluation of the complexities involved in implementing the algorithm is necessary in light of hardware and algorithm advances. In this work, we investigate the Hestenes-Jacobi SVD (HJSVD) on FPGAs. Our findings show that the Hestenes-Jacobi method, while highly parallelizable, can become constrained in its hardware resource costs and requires careful tuning to achieve high throughput with input matrix sizes and degree of parallelism having an important effect on efficient pipelining of the architecture. We identify the key challenges in parallelizing the algorithm for modern workloads and incorporate three key optimizations: pipelining, the use of fixed-point arithmetic instead of floating-point, and the use of heterogeneous resources for vector rotations, not only DSPs.

      November 2023  31st European Signal Processing Conference (EUSIPCO) Conference

    Accelerated Sample-Accurate R-Peak Detectors Based on Visibility Graphs

    Jonas Emrich, Taulant Koka, Sebastian Wirth, Michael Muma

    BibTeX DOI: 10.23919/EUSIPCO58844.2023.10290007

    Abstract
    The effective detection and accurate clinical diagnosis of cardiac conditions strongly relies on the correct localization of R-peaks in the electrocardiogram (ECG). Recently, demand for sample-accurate R-peak detection, which is essential to precisely reveal vital features, such as heart rate variability and pulse transit time, has increased. Therefore, we propose two novel sample-accurate visibility-graph-based R-peak detectors, the FastNVG and the FastWHVG detector. The visibility graph (VG) transformation maps a discrete signal into a graph by representing sampling locations as nodes and establishing edges between mutually visible samples. However, processing large-scale clinical ECG data urgently demands further acceleration of VG-based algorithms. The proposed methods reduce the required computation time by one order of magnitude and simultaneously decrease the required memory compared to a recently proposed VG-based R-Peak detector. Instead of transforming the entire ECG, the proposed acceleration benefits largely from building the VG based on a subset containing only the samples relevant to R-peak detection. Further acceleration is obtained by adopting the computationally efficient horizontal visibility graph, which has not yet been used for R-peak detection. Numerical experiments and benchmarks on multiple ECG databases demonstrate a significantly superior performance of the proposed VG-based methods compared to popular R-peak detectors.

      November 2023  IEEE Communications Letters Article

    Glucose Regulation Through Cooperative Molecular Communication

    Theodoros M. Theodoridis, Sotiris A. Tegos, Panagiotis D. Diamantoulakis, Vahid Jamali, George K. Karagiannidis

    BibTeX DOI: 10.1109/LCOMM.2023.3315012

    Abstract
    In Type 1 diabetes, the pancreatic beta cells responsible for producing insulin are destroyed by the immune system. Insulin is needed to activate an insulin-dependent glucose transporter, which is responsible for taking glucose into the muscle cell for metabolism. Recent advances in nanotechnology, bioengineering and synthetic biology are bringing the artificial beta cell (ABC) closer to reality. In this letter, we model glucose regulation by ABCs as a cooperative molecular communication system, in which the glucose source is seen as the transmitter and the muscle as the receiver. the last absorbs the glucose in the presence of insulin, and the ABC is modeled as a a decode-and-forward relay that detects glucose molecules and releases insulin in response. Using this model, we analyze the end-to-end system performance for ABC-assisted glucose regulation by providing closed-form expressions for the probabilities of hyperglycemia and hypoglycemia and the error probability of the system. In addition, we present simulation results for quantifying performance and validation of the analysis.

      November 2023  31st European Signal Processing Conference (EUSIPCO 2023) Conference

    Sparsity-Aware Block Diagonal Representation for Subspace Clustering

    Aylin Taştan, Michael Muma, Esa Ollila, Abdelhak M. Zoubir

    BibTeX DOI: 10.23919/EUSIPCO58844.2023.10289969

    Abstract
    A block diagonally structured affinity matrix is an informative prior for subspace clustering which embeds the data points in a union of low-dimensional subspaces. Structuring a block diagonal matrix can be challenging due to the determination of an appropriate sparsity level, especially when outliers and heavy-tailed noise obscure the underlying subspaces. We propose a new sparsity-aware block diagonal representation (SABDR) method that robustly estimates the appropriate sparsity level by leveraging upon the geometrical analysis of the low-dimensional structure in spectral clustering. Specifically, we derive the Euclidean distance between the embeddings of different clusters to develop a computationally efficient density-based clustering algorithm. In this way, the sparsity parameter selection problem is re-formulated as a robust approximation of target between-clusters distances. Comprehensive experiments using real-world data demonstrate the effectiveness of SABDR in different subspace clustering applications.

      November 2023  2023 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR’23) Conference

    Accurate Pose Prediction on Signed Distance Fields for Mobile Ground Robots in Rough Terrain

    Martin Oehler, Oskar von Stryk

    PDF BibTeX DOI: 10.1109/SSRR59696.2023.10499944

    Abstract
    Autonomous locomotion for mobile ground robots in unstructured environments such as waypoint navigation or flipper control requires a sufficiently accurate prediction of the robot-terrain interaction. Heuristics like occupancy grids or traversability maps are widely used but limit actions available to robots with active flippers as joint positions are not taken into account. We present a novel iterative geometric method to predict the 3D pose of mobile ground robots with active flippers on uneven ground with high accuracy and online planning capabilities. This is achieved by utilizing the ability of signed distance fields to represent surfaces with sub-voxel accuracy. The effectiveness of the presented approach is demonstrated on two different tracked robots in simulation and on a real platform. Compared to a tracking system as ground truth, our method predicts the robot position and orientation with an average accuracy of 3.11 cm and 3.91°, outperforming a recent heightmap-based approach. The implementation is made available as an open-source ROS package.

      October 2023  Conference on Communications and Network Security (CNS 2023) Conference

    BeamSec: A Practical mmWave Physical Layer Security Scheme Against Strong Adversaries

    Afifa Ishtiaq, Arash Asadi, Ladan Khaloopour, Waqar Ahmed, Vahid Jamali, Matthias Hollick

    BibTeX DOI: 10.1109/CNS59707.2023.10289003

    Abstract
    The high directionality of millimeter-wave (mmWave) communication systems has proven effective in reducing the attack surface against eavesdropping, thus improving the physical layer security. However, even with highly directional beams, the system is still exposed to eavesdropping against adversaries located within the main lobe. In this paper, we propose BeamSec, a solution to protect the users even from adversaries located in the main lobe. The key feature of BeamSec are: (i) Operating without the knowledge of eavesdropper’s location/channel; (ii) Robustness against colluding eavesdropping attack and (iii) Standard compatibility, which we prove using experiments via our IEEE 802.11ad/ay-compatible 60 GHz phased-array testbed. Methodologically, BeamSec first identifies uncorrelated and diverse beampairs between the transmitter and receiver by analyzing signal characteristics available through standard-compliant procedures. Next, it encodes the information jointly over all selected beampairs to minimize information leakage. We study two methods for allocating transmission time among different beams, namely uniform allocation (no knowledge of the wireless channel) and optimal allocation for maximization of the secrecy rate (with partial knowledge of the wireless channel). Our experiments show that BeamSec outperforms the benchmark schemes against single and colluding eavesdroppers and enhances the secrecy rate by 79.8% over a random paths selection benchmark.

      October 2023  IEEE International Conference on Communication Conference

    Completion Time Minimization for UAV-Based Communications with a Finite Buffer

    Yi Wang, Lin Xiang, Anja Klein

    BibTeX DOI: 10.1109/ICC45041.2023.10278804

    Abstract
    This paper considers a buffer-aided unmanned aerial vehicle (UAV) serving as an aerial relay for communication between a base station (BS) and multiple ground users (GUs). Thanks to its flexible mobility, the UAV can achieve high-rate communications with the GUs/BS by buffering the communication data and exploiting the favorable channel conditions on its flight trajectory for transmission and reception. However, the size of the buffer is limited in practice, which may severely restrict the throughput gains enabled by buffering. Whether it is beneficial to consider buffering at UAVs with a small buffer is an open research problem, which is tackled in this paper. Assuming a finite buffer mounted at the UAV, we consider joint optimization of the resource allocation, data buffering, and trajectory planning for minimizing the UAV’s completion time required for delivering a given data volume from each GU to the BS, where the resource allocation contains power and bandwidth allocation. The formulated optimization problem is a mixed-integer nonconvex program, which is generally intractable. To solve this problem, we propose a novel low-complexity two-layer iterative suboptimal algorithm based on bisection search and penalty successive convex approximation (PSCA). Note that minimizing the completion time in turn maximizes the average throughput, i.e., the amount of data delivered from the GUs to the BS per unit of time. Simulation results show that the buffer with sufficiently large size can increase the UAV’s average throughput by up to 123.8% compared to without buffering. Moreover, with our proposed scheme, 63.2% of the throughput gains can already be achieved using only a small buffer.

      October 2023  58th International Conference on Communications Conference

    Online Learning in Matching Games for Task Offloading in Multi-Access Edge Computing

    Helena Mehler, Bernd Simon, Anja Klein

    BibTeX DOI: 10.1109/ICC45041.2023.10279031

    Abstract
    In multi-access edge computing (MEC), mobile users (MUs) can offload computation tasks to nearby computational resources, which are owned by a mobile network operator (MNO), to save energy. In this work, we investigate two important challenges of task offloading in MEC: (i) The techno-economic interactions of the MNO and the MUs. The MNO faces a profit maximization problem, whereas the MUs face an energy minimization problem. (ii) Limited information at the MUs about the MNO’s communication and computation resources and the task offloading strategies of other MUs. To overcome these challenges, we model the task offloading problem as a matching game between the MUs and the MNO including their techno-economic interactions. Furthermore, we propose a novel Collision-Avoidance Task Offloading Multi-Armed-Bandit (CA-TO-MAB) algorithm, that allows the MUs to learn the amount of available resources at the MNO and the task offloading strategies of other MUs in an online, fully decentralized way. We show that by using CA-TO-MAB, the cumulative revenue of the MNO can be increased by 25% and, at the same time the energy consumption of the MUs can be reduced by 6% compared to state-of-the-art online learning algorithms for task offloading. Furthermore, the communication overhead can be reduced by 55% compared to a non-learning game-theoretic approach.

      October 2023  2023 IEEE International Conference on Communications Workshop Conference

    Risk-Sensitive Optimization and Learning for Minimizing Age of Information in Point-to-Point Wireless Communications

    Wanja De Sombre, Andrea Patricia Ortiz Jimenez, Frank Aurzada, Anja Klein

    BibTeX DOI: 10.1109/ICCWorkshops57953.2023.10283567

    Abstract
    When using Internet of Things (IoT) networks for monitoring, devices rely on fresh status updates about the monitored process. To measure the freshness of these status updates, the concept of Age of Information (AoI) is used. However, critical applications, e.g., those involving human safety, require not only fresh updates, but also a low risk of experiencing high AoI values. In this work, we introduce the notion of risky states for these high AoI events. We consider a point-to-point wireless communication scenario containing a sender transmitting randomly arriving status updates to a receiver through a wireless channel. The sender decides, when to send a status update and when to wait for a newer one. The sender’s goal is to jointly minimize the AoI at the receiver, the required transmission energy and the frequency of visiting risky states. We present two solutions for this problem using optimization and learning, respectively For the optimization approach, we propose a family of threshold-based transmission strategies, which trigger a transmission whenever the difference between the AoI at the sender and at the receiver exceeds a certain threshold. Our proposed learning approach directly includes our notion of risky states into traditional Q -learning As a result, it balances the minimization of AoI and the required transmission energy, with the frequency of visiting risky states. Through numerical results, we show that our proposed risk-aware approaches outperform relevant reference schemes. Moreover, and in contrast to value iteration, their computational complexity does not depend on the set of possible AoI values.

      October 2023  ICC 2023 - IEEE International Conference on Communications Conference

    Deterministic Identification for MC ISI-Poisson Channel

    Mohammad J. Salariseddigh, Vahid Jamali, Uzi Pereg, Holger Boche, Christian Deppe, Robert Schober

    BibTeX DOI: 10.1109/ICC45041.2023.10278856

    Abstract
    Several applications of molecular communications (MC) feature an alarm-prompt behavior for which the prevalent Shannon capacity may not be the appropriate performance metric. The identification capacity as an alternative measure for such systems has been motivated and established in the literature. In this paper, we study deterministic identification (DI) for the discrete-time Poisson channel (DTPC) with intersymbol interference (ISI) where the transmitter is restricted to an average and a peak molecule release rate constraint. Such a channel serves as a model for diffusive MC systems featuring long channel impulse responses and employing molecule counting receivers. We derive lower and upper bounds on the DI capacity of the DTPC with ISI when the number of ISI channel taps K may grow with the codeword length n (e.g., due to increasing symbol rate). As a key finding, we establish that for deterministic encoding, the codebook size scales as 2(nlogn)R assuming that the number of ISI channel taps scales as K=2κlogn , where R is the coding rate and κ is the ISI rate. Moreover, we show that optimizing κ leads to an effective identification rate [bits/s] that scales linearly with n , which is in contrast to the typical transmission rate [bits/s] that is independent of n .

      October 2023  Weizenbaum Journal of the Digital Society Article

    Digital Volunteers during the COVID-19 Pandemic: Care Work on Social Media for Socio-technical Resilience

    Stefka Schmid, Laura Gianna Guntrum, Steffen Haesler, Lisa Schultheiß, Christian Reuter

    BibTeX DOI: 10.34669/WI.WJDS/3.3.6

    Abstract
    Like past crises, the COVID-19 pandemic has galvanized individual volunteers to contribute to the public response. This includes digital volunteers who have organized physical aid and conducted social media activities. Analyzing German volunteering support groups on Facebook and related Reddit threads in the context of COVID-19, we show what types of help are offered and how social media users interact with each other to cope with the situation. We reveal that most users offering help online mostly perform typical care work, such as buying groceries or giving advice. Crucially, volunteering is characterized by relationships of care. This means it builds on affirmative interactions. In spite of some misdirected offers and regressive interruptions, people use the possibility to make their voices heard and, showing empathy, help each other to live with the crisis. Social media like Facebook mediate societal structures, including relationships of care, offering a space for the continuous, cumulatively resilient conduct of care work. Reflecting on the traditional division of labor in crisis volunteering and counter-productive dynamics of care and empathy, we aim to articulate a feminist ethics of care that allows for interactions on social media that foster generative computer-supported collaboration.

      October 2023  11th ACM Symposium on Spatial User Interaction Conference

    DensingQueen: Exploration Methods for Spatial Dense Dynamic Data

    Julius von Willich, Sebastian Günther, Andrii Matviienko, Martin Schmitz, Florian Müller, Max Mühlhäuser

    PDF BibTeX DOI: 10.1145/3607822.3614535

    Abstract
    Research has proposed various interaction techniques to manage the occlusion of 3D data in Virtual Reality (VR), e.g., via gradual refinement. However, tracking dynamically moving data in a dense 3D environment poses the challenge of ever-changing occlusion, especially if motion carries relevant information, which is lost in still images. In this paper, we evaluated two interaction modalities for Spatial Dense Dynamic Data (SDDD), adapted from existing interaction methods for static and spatial data. We evaluated these modalities for exploring SDDD in VR, in an experiment with 18 participants. Furthermore, we investigated the influence of our interaction modalities on different levels of data density on the users’ performance in a no-knowledge task and a prior-knowledge task. Our results indicated significantly degraded performance for higher levels of density. Further, we found that our flashlight-inspired modality successfully improved tracking in SDDD, while a cutting plane-inspired approach was more suitable for highlighting static volumes of interest, particularly in such high-density environments.

      October 2023  IEEE Transactions on Molecular, Biological and Multi-Scale Communications Article

    Deterministic Identification for Molecular Communications Over the Poisson Channel

    Mohammad Javad Salariseddigh, Vahid Jamali, Uzi Pereg, Holger Boche, Christian Deppe, Robert Schober

    BibTeX DOI: 10.1109/TMBMC.2023.3324487

    Abstract
    Various applications of molecular communications (MC) are event-triggered, and, as a consequence, the prevalent Shannon capacity may not be the right measure for performance assessment. Thus, in this paper, we motivate and establish the identification capacity as an alternative metric. In particular, we study deterministic identification (DI) for the discrete-time Poisson channel (DTPC), subject to an average and a peak molecule release rate constraint, which serves as a model for MC systems employing molecule counting receivers. It is established that the number of different messages that can be reliably identified for this channel scales as 2(nlogn)R , where n and R are the codeword length and coding rate, respectively. Lower and upper bounds on the DI capacity of the DTPC are developed. The obtained large capacity of the DI channel sheds light on the performance of natural DI systems such as natural olfaction, which are known for their extremely large chemical discriminatory power in biology. Furthermore, numerical results for the empirical miss-identification and false identification error rates are provided for finite length codes. This allows us to characterize the behaviour of the error rate for increasing codeword lengths, which complements our theoretically-derived scale for asymptotically large codeword lengths.

      October 2023  29th Annual International Conference on Mobile Computing and Networking (ACM MobiCom ‘23) Conference

    Introducing FreeSpeaker - A Modular Smart Home Hub Prototyping Platform

    Hermann Leinweber, Jonatan Crystall, Frank Hessel, Florentin Putz, Matthias Hollick

    PDF BibTeX DOI: 10.1145/3570361.3614080

    Abstract
    Smart home speakers have become a commodity item in many households and provide interesting research opportunities in areas like wireless communication and human-computer interaction. Commercial devices do not provide sufficient access for many research tasks. We present a modular smart home hub designed specifically for research purposes. The electronic and mechanical components are designed with reproducibility in mind and can be easily recombined for a project’s needs. Additionally, we show applications of the hub in different scenarios.

      October 2023  29th Annual International Conference on Mobile Computing and Networking (ACM MobiCom ‘23) Conference

    Software-Defined Wireless Communication Systems for Heterogeneous Architectures

    David Volz, Andreas Koch, Bastian Bloessl

    BibTeX DOI: 10.1145/3570361.3614084

    Abstract
    Future cellular networks will be programmable and increasingly software-defined with APIs to hook into the communication stack closer and closer to the physical layer. This allows operators, for example, to plug-in third-party machine learning algorithms to optimize performance. At the same time, this flexibility implies that compute resources cannot be provisioned statically but have to be distributed dynamically during runtime. While the hardware platforms for such systems are available, we lack suitable software frameworks that help to realize such systems. In this demonstration, we present two Open Source projects that fill this gap: FutureSDR, a portable real-time signal processing framework with native support for accelerators (like GPUs and FGPAs); and IPEC, which enables fully automatic composition of multi-accelerator FPGA designs. We believe that these tools - especially in combination with each other - can provide the base for building research prototypes and allow experimentation with software-defined wireless communication systems.

      September 2023  ACM Transactions on Programming Languages and Systems Article

    Prisma: A Tierless Language for Enforcing Contract-client Protocols in Decentralized Applications

    David Richter, David Kretzler, Pascal Weisenburger, Guido Salvaneschi, Sebastian Faust, Mira Mezini

    BibTeX DOI: 10.1145/3604629

    Abstract
    Decentralized applications (dApps) consist of smart contracts that run on blockchains and clients that model collaborating parties. dApps are used to model financial and legal business functionality. Today, contracts and clients are written as separate programs—in different programming languages—communicating via send and receive operations. This makes distributed program flow awkward to express and reason about, increasing the potential for mismatches in the client-contract interface, which can be exploited by malicious clients, potentially leading to huge financial losses. In this article, we present Prisma , a language for tierless decentralized applications, where the contract and its clients are defined in one unit and pairs of send and receive actions that “belong together” are encapsulated into a single direct-style operation, which is executed differently by sending and receiving parties. This enables expressing distributed program flow via standard control flow and renders mismatching communication impossible. We prove formally that our compiler preserves program behavior in presence of an attacker controlling the client code. We systematically compare Prisma with mainstream and advanced programming models for dApps and provide empirical evidence for its expressiveness and performance.

      September 2023  31st Interdisciplinary Information Management Talks Conference

    A Large-Scale Data Collection and Evaluation Framework for Android Device Security Attributes

    Ernst Leierzopf, Michael Roland, René Mayrhofer, Florentin Putz

    PDF BibTeX DOI: 10.35011/IDIMT-2023-63

    Abstract
    Android’s fast-lived development cycles and increasing amounts of manufacturers and device models make a comparison of relevant security attributes, in addition to the already difficult comparison of features, more challenging. Most smartphone reviews only consider offered features in their analysis. Smartphone manufacturers include their own software on top of the Android Open Source Project (AOSP) to improve user experience, to add their own pre-installed apps or apps from third-party sponsors, and to distinguish themselves from their competitors. These changes affect the security of smartphones. It is insufficient to validate device security state only based on measured data from real devices for a complete assessment. Promised major version releases, security updates, security update schedules of devices, and correct claims on security and privacy of pre-installed software are some aspects, which need statistically significant amounts of data to evaluate. Lack of software and security updates is a common reason for shorter lifespans of electronics, especially for smartphones. Validating the claims of manufacturers and publishing the results creates incentives towards more sustainable maintenance and longevity of smartphones. We present a novel scalable data collection and evaluation framework, which includes multiple sources of data like dedicated device farms, crowdsourcing, and webscraping. Our solution improves the comparability of devices based on their security attributes by providing measurements from real devices.

      September 2023  48th Conference on Local Computer Networks Conference

    UAV Swarms for Joint Data Ferrying and Dynamic Cell Coverage via Optimal Transport Descent and Quadratic Assignment

    Kai Cui, Lars Baumgärtner, Mustafa Burak Yilmaz, Mengguang Li, Christian Fabian, Benjamin Becker, Lin Xiang, Maximilian Bauer, Heinz Koeppl

    BibTeX DOI: 10.1109/LCN58197.2023.10223388

    Abstract
    Both data ferrying with disruption-tolerant networking (DTN) and mobile cellular base stations constitute important techniques for UAV-aided communication in situations of crises where standard communication infrastructure is unavailable. For optimal use of a limited number of UAVs, we propose providing both DTN and a cellular base station on each UAV. Here, DTN is used for large amounts of low-priority data, while capacity-constrained cell coverage remains reserved for emergency calls or command and control. We optimize cell coverage via a novel optimal transport-based formulation using alternating minimization, while for data ferrying we periodically deliver data between dynamic clusters by solving quadratic assignment problems. In our evaluation, we consider different scenarios with varying mobility models and a wide range of flight patterns. Overall, we tractably achieve optimal cell coverage under quality-of-service costs with DTN-based data ferrying, enabling large-scale deployment of UAV swarms for crisis communication.

      September 2023  48th Conference on Local Computer Networks Conference

    Energy-efficient Broadcast Trees for Decentralized Data Dissemination in Wireless Networks

    Artur Sterz, Robin Klose, Markus Sommer, Jonas Höchst, Jakob Link, Bernd Simon, Anja Klein, Matthias Hollick, Bernd Freisleben

    PDF BibTeX DOI: 10.1109/LCN58197.2023.10223400

    Abstract
    We present a novel multi-hop data dissemination protocol for wireless networks that minimizes the total energy consumption across an entire network by minimizing the transmission power at each hop. It is based on a game-theoretic model, constructs a spanning tree topology in a decentralized manner, and is usable in practice. We evaluate the protocol via simulation and a pratical implementation on a testbed of 75 Raspberry Pis, demonstrating that a total energy reduction of up to 90% can be achieved compared to a simple broadcast protocol.

      September 2023  2023 IEEE/CIC International Conference on Communications in China Conference

    Risk-aware Online Optimization of Cross-Layer Resource Allocation in Next-Generation WLANs

    Jing Zhang, Jinke Zheng, Lin Xiang, Xiaohu Ge

    BibTeX DOI: 10.1109/ICCC57788.2023.10233666

    Abstract
    Fiber-to-the-room (FTTR) technology provides a promising solution to enable high-capacity indoor wireless communications in next-generation wireless local area networks (WLANs). In this paper, we consider online cross-layer resource allocation in the uplink of FTTR WLANs to facilitate immersive user experience. We formulate a nonconvex optimization problem for maximizing the long-term total network utility while ensuring proportional fairness among the users and guaranteeing users’ quality of service (QoS) requirements. Unlike the existing cross-layer resource allocation schemes, our approach simultaneously ensures queue stability and mitigates the risk of data loss caused by e.g. fading and interference, preemptive multiuser access, and finite buffer capacity, both achieved by limiting the entropic value-at-risk (EVaR) of the users’ data queues. By employing Lyapunov optimization and relaxation techniques, we further transform the NP-hard problem into a convex problem and solve it via convex optimization. Simulation results show that the proposed EVaR-based cross-layer resource allocation framework can effectively limit the length of data queues at the users and, at the same time, provide fairness for the throughput achieved among users.

      September 2023  22nd International Conference on Modeling and Applied Simulation Conference

    Combining TAPAS and SUMO towards crises management based on traffic data

    María López Díaz, Andrea Tundis

    PDF BibTeX DOI: 10.46354/i3m.2023.mas.005

    Abstract
    Human-made crisis and natural disasters are major concerns for the society, as they can put in a risk the life of people. Especially in urban area, which are typically highly crowded, the occurrence of a situation of dangers the consequent change of behaviors due to panics can be difficult to imagine, which affects the ability to be able to define appropriate countermeasures to mitigate the crisis itself. That’s why the use of digital solutions can be beneficial to support the analysis of such scenario and improve the understanding of the behavior in case of crisis. In this context, our paper focuses on the transportation infrastructures and proposes an integrated solution based on the combination of two simulation tools called TAPAS and SUMO to support the modelling and simulation of mobility scenario. The integrated solution is experimented in the context of city of Darmstadt (Germany) by simulating normal and change of behavior in terms of mobility and showing the deriving benefits.

      August 2023  42nd IEEE Conference on Computer Communications (INFOCOM 2023) Conference

    Safehaul: Risk-Averse Learning for Reliable mmWave Self-Backhauling in 6G Networks

    Amir Ashtari Gargari, Andrea Patricia Ortiz Jimenez, Matteo Pagin, Anja Klein, Matthias Hollick, Michele Zorzi, Arash Asadi

    BibTeX DOI: 10.1109/INFOCOM53939.2023.10228969

    Abstract
    Wireless backhauling at millimeter-wave frequencies (mmWave) in static scenarios is a well-established practice in cellular networks. However, highly directional and adaptive beamforming in today’s mmWave systems have opened new possibilities for self-backhauling. Tapping into this potential, 3GPP has standardized Integrated Access and Backhaul (IAB) allowing the same base station to serve both access and backhaul traffic. Although much more cost-effective and flexible, resource allocation and path selection in IAB mmWave networks is a formidable task. To date, prior works have addressed this challenge through a plethora of classic optimization and learning methods, generally optimizing a Key Performance Indicator (KPI) such as throughput, latency, and fairness, and little attention has been paid to the reliability of the KPI. We propose Safehaul, a risk-averse learning-based solution for IAB mmWave networks. In addition to optimizing average performance, Safehaul ensures reliability by minimizing the losses in the tail of the performance distribution. We develop a novel simulator and show via extensive simulations that Safehaul not only reduces the latency by up to 43.2% compared to the benchmarks, but also exhibits significantly more reliable performance, e.g., 71.4% less variance in achieved latency.

      August 2023  2023 IEEE International Symposium on Information Theory (ISIT) Conference

    Deterministic Identification for MC Binomial Channel

    Mohammad Javad Salariseddigh, Vahid Jamali, Holger Boche, Christian Deppe, Robert Schober

    BibTeX DOI: 10.1109/ISIT54713.2023.10206627

    Abstract
    The Binomial channel serves as a fundamental model for molecular communication (MC) systems employing molecule-counting receivers. Here, deterministic identification (DI) is addressed for the discrete-time Binomial channels (DTBC), subject to an average and a peak constraint on the molecule release rate. We establish that the number of different messages that can be reliably identified for the DTBC scales as 2 (n log n)R , where n and R are the codeword length and coding rate, respectively. Lower and upper bounds on the DI capacity of the DTBC are developed.

      August 2023  97th Vehicular Technology Conference Conference

    Full-Link AoI Analysis of Uplink Transmission in Next-Generation FTTR WLAN

    Jing Zhang, Jing Liu, Lin Xiang, Xiaohu Ge

    BibTeX DOI: 10.1109/VTC2023-Spring57618.2023.10200633

    Abstract
    Fiber-to-the-room (FTTR) wireless local area networks (WLANs) are a promising sixth-generation (6G) technology for extreme broadband low-latency indoor wireless communications. With dense deployment of access points (APs), namely optical network units (ONUs), and efficient spatial frequency reuse across the ONUs, FTTR WLANs enable the mobile devices to flexibly access any ONU in its communication range and reduce the collisions during data packet transmissions. However, FTTR WLANs share a passive optical network (PON) for time division multiplexing (TDM) based backhauling, which may incur long delays for scheduling packet transmissions over the PON. In this paper, the full-link age of information (FL-AoI) is proposed as a new performance metric to analyze the timeliness of indoor communications in FTTR WLANs, taking into account both the carrier sense multiple access with collision avoidance (CSMA/CA) based wireless transmission and the TDM based packet scheduling over the PON. The FL-AoI of FTTR WLANs is analyzed using stochastic geometry and stretched exponential path-loss (SEPL) based indoor wireless channel model. We show that rather than accessing the nearest ONUs, the FTTR WLAN also enable the mobile devices to access further ONUs to reduce the average FL-AoI. Meanwhile, there exists an optimal transmission distance to achieve minimal average FL-AoI in the FTTR WLAN, whose value depends on the deployment density of ONUs.

      August 2023  22nd IEEE Statistical Signal Processing Workshop Conference

    False Discovery Rate Control for Fast Screening of Large-Scale Genomics Biobanks

    Jasin Machkour, Michael Muma, Daniel P. Palomar

    BibTeX DOI: 10.1109/SSP53291.2023.10207957

    Abstract
    Genomics biobanks are information treasure troves with thousands of phenotypes (e.g., diseases, traits) and millions of single nucleotide polymorphisms (SNPs). The development of methodologies that provide reproducible discoveries is essential for the understanding of complex diseases and precision drug development. Without statistical reproducibility guarantees, valuable efforts are spent on researching false positives. Therefore, scalable multivariate and high-dimensional false discovery rate (FDR)-controlling variable selection methods are urgently needed, especially, for complex polygenic diseases and traits. In this work, we propose the Screen-T-Rex selector, a fast FDR-controlling method based on the recently developed T-Rex selector. The method is tailored to screening large-scale biobanks and it does not require choosing additional parameters (sparsity parameter, target FDR level, etc). Numerical simulations and a real-world HIV-1 drug resistance example demonstrate that the performance of the Screen-T-Rex selector is superior, and its computation time is multiple orders of magnitude lower compared to current benchmark knockoff methods.

      July 2023  IEEE Intelligent Vehicles Symposium (IV 2023) Conference

    Towards Continual Knowledge Learning of Vehicle CAN-data

    Sajeel Ahmed, Ousama Esbel, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1109/iv55152.2023.10186715

    Abstract
    In this paper, we propose a continual learning (CL) approach that adapts to the vehicle CAN-data flexibly and continuously. Our approach is capable of learning from vehicle CAN-bus data in multiple driving scenarios, adapting to the various drifts within each driving scenario. The basis for our approach corresponds to a common solver model and a series of supervisor models. Our solver model extends the memory-aware synapses approach with the use of weight cloning and weighted experience replay. Our supervisor model selects the output of the solver model that corresponds to the driving scenario present at the input. We evaluate our approach using a Tesla Model 3 CAN-data and 8 different driving scenarios. Our evaluation results show that our approach effectively learns multiple driving scenarios sequentially without forgetting the previous knowledge.

      July 2023  2023 Designing Interactive Systems Conference Conference

    Getting the Residents’ Attention: The Perception of Warning Channels in Smart Home Warning Systems

    Steffen Haesler, Marc Wendelborn, Christian Reuter

    BibTeX DOI: 10.1145/3563657.3596076

    Abstract
    About half a billion households are expected to use smart home systems by 2025. Although many IoT sensors, such as smoke detectors or security cameras, are available and governmental crisis warning systems are in place, little is known about how to warn appropriately in smart home environments. We created a Raspberry Pi based prototype with a speaker, a display, and a connected smart light bulb. Together with a focus group, we developed a taxonomy for warning messages in smart home environments, dividing them into five classes with different stimuli. We evaluated the taxonomy using the Experience Sampling Method (ESM) in a field study at participants’ (N = 13) homes testing 331 warnings. The results show that taxonomy-based warning stimuli are perceived to be appropriate and participants could imagine using such a warning system. We propose a deeper integration of warning capabilities into smart home environments to enhance the safety of citizens.

      July 2023  8th Workshop on Representation Learning for NLP (RepL4NLP 2023) Conference

    One does not fit all! On the Complementarity of Vision Encoders for Vision and Language Tasks

    Gregor Geigle, Chen Liu, Jonas Pfeiffer, Iryna Gurevych

    PDF BibTeX DOI: 10.18653/v1/2023.repl4nlp-1.9

    Abstract
    Current multimodal models, aimed at solving Vision and Language (V+L) tasks, predominantly repurpose Vision Encoders (VE) as feature extractors. While many VEs—of different architectures, trained on different data and objectives—are publicly available, they are not designed for the downstream V+L tasks. Nonetheless, most current work assumes that a single pre-trained VE can serve as a general-purpose encoder. In this work, we focus on analysis and aim to understand whether the information stored within different VEs is complementary, i.e. if providing the model with features from multiple VEs can improve the performance on a target task, and how they are combined. We exhaustively experiment with three popular VEs on six downstream V+L tasks and analyze the attention and VE-dropout patterns. Our analyses suggest that diverse VEs complement each other, resulting in improved downstream V+L task performance, where the improvements are not due to simple ensemble effects (i.e. the performance does not always improve when increasing the number of encoders). We demonstrate that future VEs, which are not repurposed, but explicitly designed for V+L tasks, have the potential of improving performance on the target V+L tasks.

      July 2023  24th International Conference on Digital Signal Processing Conference

    Radar Based Humans Localization with Compressed Sensing and Sparse Reconstruction

    Christian Eckrich, Christian A. Schroth, Vahid Jamali, Abdelhak M. Zoubir

    BibTeX DOI: 10.1109/DSP58604.2023.10167990

    Abstract
    Localization and detection is a vital task in emergency rescue operations. Devastating natural disasters can create environments that are inaccessible or dangerous for human rescuers. Contaminated areas or buildings in danger of collapsing can be searched by rescue robots which are equipped with diverse sensors such as optical and radar sensors. In scenarios where the line of sight is blocked, e.g., by a wall, a door or heavy smoke or dust, sensors like LiDAR or cameras are not able to provide sufficient information. The usage of radar in these kinds of situations can drastically improve situational awareness and hence the likelihood of rescue. In this paper, we present a method that is used for radar imaging behind obstacles by utilizing a signal model that includes the floor reflection propagation path in addition to the direct path of the radar signal. Additionally, compressed sensing methods are presented and applied to real world radar data that was recorded by a Stepped Frequency Continuous Wave (SFCW) radar mounted on a semi-autonomous robot. The results show an improved radar image that allows the clear identification of persons behind obstacles.

      July 2023  24th International Conference on Digital Signal Processing Conference

    Attacks on Robust Distributed Learning Schemes via Sensitivity Curve Maximization

    Christian A. Schroth, Stefan Vlaski, Abdelhak M. Zoubir

    BibTeX DOI: 10.1109/DSP58604.2023.10167919

    Abstract
    Distributed learning paradigms, such as federated or decentralized learning, allow a collection of agents to solve global learning and optimization problems through limited local interactions. Most such strategies rely on a mixture of local adaptation and aggregation steps, either among peers or at a central fusion center. Classically, aggregation in distributed learning is based on averaging, which is statistically efficient, but susceptible to attacks by even a small number of malicious agents. This observation has motivated a number of recent works, which develop robust aggregation schemes by employing robust variations of the mean. We present a new attack based on sensitivity curve maximization (SCM), and demonstrate that it is able to disrupt existing robust aggregation schemes by injecting small, but effective perturbations.

      July 2023  40th IEEE International Conference on Robotics and Automation (ICRA2023) Conference

    Scalable Task-Driven Robotic Swarm Control via Collision Avoidance and Learning Mean-Field Control

    K. Cui, M. Li, C. Fabian, H. Koeppl

    BibTeX DOI: 10.1109/ICRA48891.2023.10161498

    Abstract
    In recent years, reinforcement learning and its multi-agent analogue have achieved great success in solving various complex control problems. However, multi-agent rein-forcement learning remains challenging both in its theoretical analysis and empirical design of algorithms, especially for large swarms of embodied robotic agents where a definitive toolchain remains part of active research. We use emerging state-of-the-art mean-field control techniques in order to convert many-agent swarm control into more classical single-agent control of distributions. This allows profiting from advances in single-agent reinforcement learning at the cost of assuming weak interaction between agents. However, the mean-field model is violated by the nature of real systems with embodied, physically colliding agents. Thus, we combine collision avoidance and learning of mean-field control into a unified framework for tractably designing intelligent robotic swarm behavior. On the theoretical side, we provide novel approximation guarantees for general mean-field control both in continuous spaces and with collision avoidance. On the practical side, we show that our approach outperforms multi-agent reinforcement learning and allows for decentralized open-loop application while avoiding collisions, both in simulation and real UAV swarms. Overall, we propose a framework for the design of swarm behavior that is both mathematically well-founded and practically useful, enabling the solution of otherwise intractable swarm problems.

      July 2023  40th IEEE International Conference on Robotics and Automation (ICRA2023) Conference

    PointCloudLab: An Environment for 3D Point Cloud Annotation with Adapted Visual Aids and Levels of Immersion

    Achref Doula, Tobias Güdelhöfer, Andrii Matviienko, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1109/icra48891.2023.10160225

    Abstract
    The annotation of 3D point cloud datasets is an expensive and tedious task. To optimize the annotation process, recent works have proposed the use of environments with higher levels of immersion in combination with different types of visual aids. However, two problems remain unresolved. First, the proposed environments limit the user to a unique level of immersion and a fixed hardware setup. Second, their design overlooks the interaction effects between the level of immersion and the visual aids on the quality of the annotation process. To address these issues, we propose PointCloudLab, an environment for 3D point cloud annotation that allows the use of different levels of immersion that work in combination with visual aids. Using PointCloudLab, we conducted a controlled experiment (N=20) to investigate the effects of levels of immersion and visual aids on the annotation process. Our findings reveal that higher levels of immersion combined with object-based visual aids lead to a faster and more accurate annotation. Furthermore, we found significant interaction effects between the levels of immersion and the visual aids on the accuracy of the annotation.

      July 2023  2023 American Control Conference Conference

    Estimating Parameter Regions for Structured Parameter Tuning via Reduced Order Subsystem Models

    Roland Schurig, Andreas Himmel, Amer Mešanović, Richard D. Braatz, Rolf Findeisen

    BibTeX DOI: 10.23919/ACC55779.2023.10156542

    Abstract
    Many large-scale systems are composed of subsystems operated by decentralized controllers, which are fixed in their structure, yet have parameters to tune. Initial tuning or subsequent adjustments dof those parameters ue to varying operating conditions or changes in the network of interconnected systems, while ensuring stability, performance, and security, pose a challenging task due to the overall complexity and size. Subsystems may not be willing or allowed to expose detailed information for safety and privacy reasons. In some cases, a comprehensive system model might not be available for global tuning, or the resulting problem might be computationally infeasible. To enable meaningful global parameter tuning while allowing for data privacy and security, we propose that the subsystems themselves should provide reduced-order models. These models capture the parametric dependency of the subsystem dynamics on the controller parameters. Specifically, we present a method to construct a region in the subsystems’ parameter space in which the deviation of the subsystem and the reduced-order model stays below a specified error bound and in which both systems are stable. A necessary and sufficient condition for such regions is derived using robust control theory. Notably, sufficiency can be expressed in terms of a linear matrix inequality. We demonstrate the approach by considering the temperature control of a large-scale building complex.

      June 2023  IEEE Control Systems Letters Article

    Towards Grassmannian Dimensionality Reduction in MPC

    Roland Schurig, Andreas Himmel, Rolf Findeisen

    PDF BibTeX DOI: 10.1109/LCSYS.2023.3291229

    Abstract
    Model predictive control presents remarkable potential for the optimal control of dynamic systems. However, the necessity for an online solution to an optimal control problem often renders it impractical for control systems with limited computational capabilities. To address this issue, specialized dimensionality reduction techniques designed for optimal control problems have been proposed. In this paper, we introduce a methodology for designing a low-dimensional subspace that provides an ideal representation for a predefined finite set of high-dimensional optimizers. By characterizing the subspace as an element of a specific Riemannian manifold, we leverage the unique geometric structure of the subspace. Subsequently, the optimal subspace is identified through optimization on the Riemannian manifold. The dimensionality reduction for the model predictive control scheme is achieved by confining the search space to the optimized low-dimensional subspace, enhancing both efficiency and applicability.

      June 2023  IEEE Nanotechnology Magazine Article

    Experimental Research in Synthetic Molecular Communications – Part I

    Sebastian Lotter, Lukas Brand, Vahid Jamali, Maximilian Schäfer, Helene M. Loos, Harald Unterweger, Sandra Greiner, Jens Kirchner, Christoph Alexiou, Dietmar Drummer, Georg Fischer, Andrea Buettner, Robert Schober

    BibTeX DOI: 10.1109/MNANO.2023.3262100

    Abstract
    Since its emergence from the communication engineering community around one and a half decades ago, the field of Synthetic Molecular Communication (SMC) has experienced continued growth, both in the number of technical contributions from a vibrant community and in terms of research funding. Throughout this process, the vision of SMC as a novel, revolutionary communication paradigm has constantly evolved, driven by feedback from theoretical and experimental studies, respectively. It is believed that especially the latter ones will be crucial for the transition of SMC towards a higher technology readiness level in the near future. In this spirit, we present here a comprehensive survey of experimental research in SMC. In particular, this survey focuses on highlighting the major drivers behind different lines of experimental research in terms of the respective envisioned applications. This approach allows us to categorize existing works and identify current research gaps that still hinder the development of practical SMC-based applications. Our survey consists of two parts: this paper and a companion paper. While the companion paper focuses on SMC with relatively long communication ranges, this paper covers SMC over short distances of typically not more than a few millimeters.

      May 2023  Energies Article

    Robust Placement and Control of Phase-Shifting Transformers Considering Redispatch Measures

    Allan Santos, Florian Steinke

    BibTeX DOI: 10.3390/en16114438

    Abstract
    Flexible AC transmission systems (FACTSs) can maximize capacity utilization under time-varying grid usage patterns by actively controlling the power flow of the transmission lines, e.g., with phase-shifting transformers (PST). In this paper, we propose an algorithm to determine the minimum number of PSTs and their location such that the grid can operate robustly for any realization of the (active) power set points from a known, continuous uncertainty set. As we show in our experiments, only considering a few extreme grid scenarios cannot provide this guarantee. The proposed algorithm considers the trade-offs between PST placement and operational decisions, such as PST control and redispatch. By minimizing the worst-case redispatch cost, it yields two affine linear control policies for these as a byproduct. Power flow is modeled as a constrained linear system, and the control design and actuator minimization tasks are formulated as a mixed-integer linear program (MILP). We also design a greedy algorithm, whose optimal value differs less than 20% from the MILP solution while being one to two orders of magnitude faster to compute. The proposed algorithm is evaluated for a small demonstrative 3-bus example and the IEEE 39 bus test system.

      May 2023  17th European Conference on Antennas and Propagation (EuCAP 2023) Conference

    Impact of Channel Models on Performance Characterization of RIS-Assisted Wireless Systems

    Vahid Jamali, Walid Ghanem, Robert Schober, H. Vincent Poor

    BibTeX DOI: 10.23919/EuCAP57121.2023.10133758

    Abstract
    The performance characterization of communication systems assisted by large reconfigurable intelligent surfaces (RISs) significantly depends on the adopted models for theunderlying channels. Under unrealistic channel models, the system performance may be over- or underestimated which yields inaccurate conclusions for the system design. In this paper, we review five channel models that are chosen to progressively improve the modeling accuracy for large RISs. For each channel model, we highlight the underlying assumptions, its advantages, and its limitations. We compare the system performance under the aforementioned channel models using RIS configuration algorithms from the literature and a new scalable algorithm proposed in this paper specifically for the configuration of extremely large RISs.

      May 2023  22nd International Conference on Information Processing in Sensor Networks (IPSN’23) Conference

    Hydra: Concurrent Coordination for Fault-tolerant Networking

    Andreas Biri, Reto Da Forno, Tobias Kuonen, Fabian Mager, Marco Zimmerling, Lothar Thiele

    BibTeX DOI: 10.1145/3583120.3587047

    Abstract
    Low-power wireless networks have the potential to enable applications that are of great importance to industry and society. However, existing network protocols do not meet the dependability requirements of many scenarios as the failure of a single node or link can completely disrupt communication and take significant time and energy to recover. This paper presents Hydra, a low-power wireless protocol that guarantees robust communication despite arbitrary node and link failures. Unlike most existing deterministic protocols, Hydra steers clear of centralized coordination to avoid a single point of failure. Instead, all nodes are equivalent in terms of protocol logic and configuration, performing coordination tasks such as synchronization and scheduling concurrently. This concept of concurrent coordination relies on a novel distributed consensus algorithm that yields provably unique decisions with low delay and energy overhead. In addition to a theoretical analysis, we evaluate Hydra in a multi-hop network of 23 nodes. Our experiments demonstrate that Hydra withstands random node failures without increasing coordination overhead and that it re-establishes efficient and reliable data exchange within seconds after a major disruption.

      May 2023  22nd International Conference on Information Processing in Sensor Networks Conference

    Demo Abstract: Building Battery-free Devices with Riotee

    Kai Geissdörfer, Ingmar Splitt, Marco Zimmerling

    PDF BibTeX DOI: 10.1145/3583120.3589808

    Abstract
    Battery-free devices eliminate the need for batteries, which are expensive, environmentally harmful, and require frequent replacement, thus reducing waste and making devices more cost-effective. We introduce Riotee, the next-generation platform for the battery-free Internet of Things. The platform comprises a base module, a debug probe that allows to conveniently update the firmware on the base module, and a number of expansion boards that extend the capabilities of the platform without the need to design a custom printed circuit board (PCB). We provide a brief overview of Riotee, and describe a demo setup that showcases the key functionality and how to get started with the platform in less than three minutes.

      May 2023  17th Conference of the European Chapter of the Association for Computational Linguistics Conference

    Delving Deeper into Cross-lingual Visual Question Answering

    Chen Liu, Jonas Pfeiffer, Anna Korhonen, Ivan Vulic, Iryna Gurevych

    PDF BibTeX DOI: 10.18653/v1/2023.findings-eacl.186

    Abstract
    Visual question answering (VQA) is one of the crucial vision-and-language tasks. Yet, existing VQA research has mostly focused on the English language, due to a lack of suitable evaluation resources. Previous work on cross-lingual VQA has reported poor zero-shot transfer performance of current multilingual multimodal Transformers with large gaps to monolingual performance, without any deeper analysis. In this work, we delve deeper into the different aspects of cross-lingual VQA, aiming to understand the impact of 1) modeling methods and choices, including architecture, inductive bias, fine-tuning; 2) learning biases: including question types and modality biases in cross-lingual setups. The key results of our analysis are: 1. We show that simple modifications to the standard training setup can substantially reduce the transfer gap to monolingual English performance, yielding +10 accuracy points over existing methods. 2. We analyze cross-lingual VQA across different question types of varying complexity for different multilingual multimodal Transformers, and identify question types that are the most difficult to improve on. 3. We provide an analysis of modality biases present in training data and models, revealing why zero-shot performance gaps remain for certain question types and languages.

      May 2023  44th Annual Conference of the European Assocoation for Computer Graphics (EUROGRAPHICS 2023) Conference

    Quick-Pro-Build: A Web-based Approach for Quick Procedural 3D Reconstructions of Buildings.

    B. Bohlender, M. Mühlhäuser, A. Sanchez Guinea

    PDF BibTeX DOI: 10.2312/egs.20231001

    Abstract
    We present Quick-Pro-Build, a web-based approach for quick procedural 3D reconstruction of buildings. Our approach allows users to quickly and easily create realistic 3D models using two integrated reference views: street view and satellite view. We introduce a novel conditional and stochastic shape grammar to represent the procedural models based on the well-established CGA shape grammar. Based on our grammar and user interface, we propose 3 modalities for procedural modeling: 1) model from scratch, 2) copy, paste, and adapt, and 3) summarize, select and adapt. The third modality enables users to model a building by summarizing similar models into an architectural style description, selecting a model from the style description, and adapting it to the target building. Summarizing and selecting allows the third modality to be the most efficient option when modeling a building with a style similar to existing buildings. The third modality is enabled by a novel algorithm that can find and combine similarities from procedural models into a style description and allows learning the preference of the users for one model inside the style description.

      April 2023  ACM 2023 CHI Conference on Human Factors in Computing Systems Conference

    FIDO2 the Rescue? Platform vs. Roaming Authentication on Smartphones

    Leon Würsching, Florentin Putz, Steffen Haesler, Matthias Hollick

    PDF BibTeX DOI: 10.1145/3544548.3580993

    Abstract
    Modern smartphones support FIDO2 passwordless authentication using either external security keys or internal biometric authentication, but it is unclear whether users appreciate and accept these new forms of web authentication for their own accounts. We present the first lab study (N=87) comparing platform and roaming authentication on smartphones, determining the practical strengths and weaknesses of FIDO2 as perceived by users in a mobile scenario. Most participants were willing to adopt passwordless authentication during our in-person user study, but closer analysis shows that participants prioritize usability, security, and availability differently depending on the account type. We identify remaining adoption barriers that prevent FIDO2 from succeeding password authentication, such as missing support for contemporary usage patterns, including account delegation and usage on multiple clients.

      April 2023  Computer Communications Article

    Adaptive Global Coordination of Local Routing Policies for Communication Networks

    Allan Santos, Amr Rizk, Florian Steinke

    BibTeX DOI: 10.1016/j.comcom.2023.03.027

    Abstract
    We consider optimal routing of data packets in communication networks featuring time-variable flow rates and bandwidth limitations. Taking into account recent programmability developments in communication systems, we propose a two-level control scheme: routers with a programmable data plane implement local proportional control policies that forward the incoming data to different available output interfaces at line rate. The local controllers’ parameters are adapted periodically on a slower time scale by a logically centralized (software-defined) network controller running a global coordination algorithm that keeps the routing feasible and optimal with respect to a network metric, such as the average packet delay. A robust optimization approach is selected to handle traffic variations in-between global adaptation steps. The outcome is a non-convex Quadratically Constrained Quadratic Program (QCQP), for which we present an iterative solution approach that is computationally suitable for realistically-sized backbone communication networks. With simulation experiments, we demonstrate the advantages of adaptive, global routing coordination compared to fixed, globally or locally-determined policies, especially concerning packet loss.

      April 2023  IEEE Nanotechnology Magazine Article

    Experimental Research in Synthetic Molecular Communications - Part II

    Sebastian Lotter, Lukas Brand, Vahid Jamali, Maximilian Schäfer, Helene M. Loos, Harald Unterweger, Sandra Greiner, Jens Kirchner, Christoph Alexiou, Dietmar Drummer, Georg Fischer, Andrea Buettner, Robert Schober

    BibTeX DOI: 10.1109/MNANO.2023.3262377

    Abstract
    In this second part of our survey on experimental research in Synthetic Molecular Communication (SMC), we review works on long-range SMC systems, i.e., systems with communication ranges of more than a few millimeters. Despite the importance of experimental research for the evolution of SMC towards a mature communication paradigm that will eventually support revolutionary applications beyond the reach of today’s prevalent communication paradigms, the existing body of literature is still comparatively sparse. Long-range SMC systems have been proposed in the literature for information transmission in two types of fluid media, liquid and air. While both types of SMC systems, i.e., liquid-based and air-based systems, rely on encoding and transmitting information using molecules, they differ substantially in terms of the physical system designs and in the type of applications they are intended for. In this article, we present a systematic characterization of experimental works on long-range SMC that reveals the major drivers of these works in terms of the respective target applications. Furthermore, the physical designs for long-range SMC proposed in the literature are comprehensively reviewed. In this way, our survey will contribute to making experimental research in this field more accessible and identifying novel directions for future research.

      April 2023  International Journal of Human-Computer Studies Article

    Preparedness nudging for warning apps? A mixed-method study investigating popularity and effects of preparedness alerts in warning apps

    Jasmin Haunschild, Selina Pauli, Christian Reuter

    BibTeX DOI: 10.1016/j.ijhcs.2023.102995

    Abstract
    Warning apps are used by many to receive warnings about imminent disasters. However, their potential for increasing awareness about general hazards and for increasing preparedness is currently underused. With a mixed-method design that includes a representative survey of the German population, a design workshop and an app evaluation experiment, this study investigates users’ preferences regarding non-acute preparedness alerts’ inclusion in crisis apps and the effectiveness of Nudging in this context. The experiment shows that while the social influence nudge had no significant effect compared to the control group without a nudging condition, the confrontational nudge increased the number of taken recommended preparedness measures. The evaluation indicates that the preparedness alerts increased users’ knowledge and their motivation to use a warning app. This motivation is, in contrast, decreased when the messages are perceived as a disruption. While many oppose push notifications, favor finding persuasively designed preparedness advice in a separate menu or as an optional notification.

      April 2023  Journal of Systems and Software Article

    The Uphill Journey of FaaS in the Open-Source Community

    Nafise Eskandani, Guido Salvaneschi

    PDF BibTeX DOI: 10.1016/j.jss.2022.111589

    Abstract
    Since its introduction in 2014 by Amazon, the Function as a Service (FaaS) model of serverless computing has set the expectation to fulfill the promise of on-demand, pay-as-you-go, infrastructure-independent processing, originally formulated by cloud computing. Yet, serverless applications are fundamentally different than traditional service-oriented software in that they pose specific performance (e.g., cold start), design (e.g., stateless), and development challenges (e.g., debugging). A growing number of cloud solutions have been continuously attempting to address each of these challenges as a result of the increasing popularity of FaaS. Yet, the characteristics of this model have been poorly understood; therefore, the challenges are poorly tackled. In this paper, we assess the state of FaaS in open-source community with a study on almost 2K real-world serverless applications. Our results show a jeopardized ecosystem, where, despite the hype of serverless solutions in the last years, a number of challenges remain untackled, especially concerning component reuse, support for software development, and flexibility among different platforms — resulting in arguably slow adoption of the FaaS model. We believe that addressing the issues discussed in this paper may help researchers shaping the next generation of cloud computing models.

      April 2023  IEEE Transactions on Communications Article

    Olfaction-Inspired MCs: Molecule Mixture Shift Keying and Cross-Reactive Receptor Arrays

    Vahid Jamali, Helene M. Loos, Andreas Buettner, Robert Schober, H. Vincent Poor

    BibTeX DOI: 10.1109/TCOMM.2023.3242379

    Abstract
    In this paper, we propose a novel concept for engineered molecular communication (MC) systems inspired by animal olfaction. We focus on a multi-user scenario where several transmitters wish to communicate with a central receiver. We assume that each transmitter employs a unique mixture of different types of signaling molecules to represent its message and the receiver is equipped with an array comprising R different types of receptors in order to detect the emitted molecule mixtures. The design of an MC system based on orthogonal molecule-receptor pairs implies that the hardware complexity of the receiver linearly scales with the number of signaling molecule types Q (i.e., R=Q ). Natural olfaction systems avoid such high complexity by employing arrays of cross-reactive receptors, where each type of molecule activates multiple types of receptors and each type of receptor is predominantly activated by multiple types of molecules albeit with different activation strengths. For instance, the human olfactory system is believed to discriminate several thousands of chemicals using only a few hundred receptor types, i.e., Q≫R . Motivated by this observation, we first develop an end-to-end MC channel model that accounts for the key properties of olfaction. Subsequently, we present the proposed transmitter and receiver designs. In particular, given a set of signaling molecules, we develop algorithms that allocate molecules to different transmitters and optimize the mixture alphabet for communication. Moreover, we formulate the molecule mixture recovery as a convex compressive sensing problem which can be efficiently solved via available numerical solvers. Finally, we present a comprehensive set of simulation results to evaluate the performance of the proposed MC designs revealing interesting insights regarding the design parameters. For instance, we show that mixtures comprising few types of molecules are best suited for communication since they can be more reliably detected by the cross-reactive array than one type of molecule or mixtures of many molecule types.

      March 2023  IEEE Access Article

    A Survey of Trusted Computing Solutions Using FPGAs

    Paul D. Rosero-Montalvo, Zsolt István, Wilmar Hernandez

    BibTeX DOI: 10.1109/ACCESS.2023.3261802

    Abstract
    Ensuring the security and privacy of computation and data management in the cloud and edge is an ever-important requirement. There are several working solutions today for trusted computing with general purpose processors, for instance, Intel SGX and ARM TrustZone. However, with the widespread commercial adoption of specialized hardware accelerators in the cloud and at the edge, most importantly FPGAs, two questions emerge: 1) How secure are they against threats? and 2) How could FPGAs be utilized for more efficient trusted computing? In this survey, we investigate these two questions precisely. Even though there have been numerous surveys in the past on the security of FPGAs, we believe it is timely to study the space of related work again, given the large number of data-centric applications aimed at targeting trusted execution environments that have recently appeared. Therefore, in addition to presenting an overview of state of the art, we also highlight some opportunities for FPGAs in the context of providing efficient trusted computation.

      March 2023 Other

    Lessons Learned: Koordination im Katastrophenmanagement

    Michèle Knodt, Eva Platzer

    PDF BibTeX DOI: 10.5281/zenodo.7756274

    Abstract
    Das Starkregenereignis vom Sommer 2021 im Ahrtal hat die Debatte über die Bewältigung von Katastrophen und deren Folgen, wie u.a. langanhaltende Stromausfälle, ganz nach oben auf die Tagesordnung befördert. Zusammen mit den 2022 deutlich gewordenen Herausforderungen des Klimawandels und den möglichen Auswirkungen des Kriegs in der Ukraine wird verstärkt über das Verbesserungspotenzial des deutschen Bevölkerungs- und Katastrophenschutzes diskutiert. Dieses Policy Paper soll dabei einen Beitrag zur Debatte leisten. Es wird gezeigt, wie wichtig ein gut ausgebautes und organisiertes Katastrophenmanagement ist und wo aktuelle Schwachstellen liegen. Darüber hinaus werden Handlungsempfehlungen zur Verbesserung des Katastrophenmanagements gegeben. Dabei fokussieren wir uns vor allem auf Koordinationsprozesse zwischen den Beteiligten des Katastrophenmanagements in der direkten Reaktion auf das Ereignis und dessen Bewältigung. Wir werden in unserer Analyse aber auch die Konsequenzen unserer Analyse für die Vorbereitung auf zukünftige Katastrophen mit einbeziehen. Wir verstehen unsere „Lessons Learned“ in diesem Papier als Beitrag zur aktuellen Diskussion, der sich auch das kürzlich beschlossene gemeinsame Kompetenzzentrum Bevölkerungsschutz (GeKoB) annimmt. Das sich damit geöffnete „window of opportunity“ wollen wir nutzen, um Verbesserungspotenzial in der koordinierten Zusammenarbeit auf allen Ebenen im deutschen Bevölkerungs- und Katastrophenschutz aufzudecken. Unsere Handlungsempfehlungen für ein besseres Katastrophenmanagement lassen sich in vier Punkten zusammenfassen: Verbesserung der Koordination innerhalb und zwischen den Katastrophenschutz- und Verwaltungsstäben durch (a) gut vernetzte Expertenteams zur Unterstützung lokaler Katastrophenschutzstabsmitglieder, (b) Überarbeitung des Ausbildungs- und Einsatzkonzeptes der Verwaltungsstäbe und (c) Vereinheitlichung der (Fach-)Sprache im Einsatz Verbesserung der Koordination der Stäbe mit den Einsatzkräften: Transparenz und Routine fördern Verbesserung der Rolle der politisch Verantwortlichen: Ausbildung und Einbindung der politisch Verantwortlichen auf allen Ebenen Verbessertes Schnittstellenmanagement zwischen Stäben und Zivilgesellschaft: Spontanhelfer*innen als Ressource begreifen

      March 2023  Transforming Cities Article

    Krisenmanagement im Ahrtal 2021

    Eva Platzer, Michèle Knodt

    PDF BibTeX

    Abstract
    Die Überschwemmungen im Ahrtal im Sommer 2021 zeigten die Herausforderungen für den Katastrophenschutz bei der Koordination der Stäbe mit der Zivilbevölkerung, den Einsatzkräften und den politisch Verantwortlichen als zentrale Akteure für die Bewältigung eines Ereignisses. Der Beitrag zeigt die Notwendigkeit einer funktionierenden Koordination zwischen diesen Beteiligten. Auf Grundlage von Berichten und Experteninterviews werden Koordinationsprobleme identifiziert und Optionen formuliert: Verbesserte Koordination zwischen Stäben und Zivilgesellschaft sowie zwischen Stäben und Einsatzkräften und die Befähigung politisch Verantwortlicher zur Gesamtkoordination.

      March 2023  Energy Policy Article

    Power blackout: Citizens’ contribution to strengthen urban resilience

    Michèle Knodt, Anna Stöckl, Florian Steinke, Martin Pietsch, Gerrit Hornung, Jan-Philipp Stroscher

    PDF BibTeX DOI: 10.1016/j.enpol.2023.113433

    Abstract
    A long-lasting, large-scale power blackout has a huge impact on the infrastructure of public life, as well as on critical infrastructure including electricity and water supply. At the same time, it can be observed that the share of renewable energies, and thus the possibility of self-sufficiency, has increased enormously in recent years. This contribution focuses on the question to what extend citizens are willing to share their electricity resources in order to make their city more resilient. In reference to Ostrom’s concept of club or common goods, it can be shown if and how the private good of citizen’s electricity resources can be transformed into a club or even a common good. Drawing on survey data from the city of Darmstadt we investigated the willingness to share electricity and to participate in participatory formats to enhance urban resilience.

      March 2023  IEEE Transactions on Molecular, Biological and Multi-Scale Communications Article

    Chemical Reactions-Based Detection Mechanism for Molecular Communications

    Trang Ngoc Cao, Vahid Jamali, Wayan Wicke, Nikola Zlatanov, Phee Lep Yeoh, Jamie Evans, Robert Schober

    BibTeX DOI: 10.1109/TMBMC.2023.3244649

    Abstract
    In molecular communications, the direct detection of signaling molecules may be challenging due to a lack of suitable sensors and interference from co-existing substances in the environment. Motivated by research in molecular biology, we investigate an indirect detection mechanism using chemical reactions between the signaling molecules and a molecular probe to produce an easy-to-measure product at the receiver. We consider two implementations of the proposed detection mechanism, i.e., unrestricted probe movement and probes restricted to a volume around the receiver. In general, the resulting reaction-diffusion equations that describe the concentrations of the reactant and product molecules in the system are non-linear and coupled, and cannot be solved in closed form. To evaluate these molecule concentrations, we develop an efficient iterative algorithm by discretizing the time variable and solving for the space variables of the concentration equations in each time step. In the special case when the concentration of the unrestricted probes is high and not significantly changed by the chemical reaction, we obtain insightful closed-form solutions. Our results show that the concentrations of the product molecules and the signalling molecules have a similar characteristic over time, i.e., a single peak and a long tail. We highlight that by carefully choosing the molecular probe and optimizing the decision threshold, the BER can be improved significantly such that a direct detection system is outperformed. Moreover, when the molecular probes are kept in a small volume around the receiver, fewer resources are needed to achieve a lower BER and/or a higher data rate compared to the case of unrestricted molecular probes.

      February 2023  IEEE Internet of Things Journal Article

    Hybrid Anomaly Detection Model on Trusted IoT Devices

    Paul D. Rosero-Montalvo, Zsolt István, Pinar Tözün, Wilmar Hernandez

    BibTeX DOI: 10.1109/JIOT.2023.3243037

    Abstract
    Most machine learning proposals in the Internet of Things (IoT) are designed and evaluated on preprocessed data sets, where data acquisition and cleaning steps are often considered a black box. In addition, IoT environments have numerous challenges related to acquiring data from sensors, where sensitive data can be threatened by malicious users who seek to interfere with the communication channel or storage. Additionally, sensor data can also be affected by noise. Therefore, differentiating the type of threat/anomaly requires additional energy and computational resources. We propose to carry out data cleaning/anomaly techniques on the IoT device itself, not in the cloud servers but closer to the data source. Therefore, the IoT device sends trusted data to the Cloud. Among the benefits of this is the considerable reduction in the cost of implementation due to less movement of data between IoT devices and the Cloud. Consequently, we define three anomaly detection steps using smoothing filters, unsupervised learning, and deep learning techniques (i.e., hybrid model) to detect different variations of anomalies and threats while focusing on a small computational/memory footprint. The deployment of the hybrid model on AVR, Tensilica, and ARM microcontrollers showed that the last ones are an adequate target to implement the model because they best satisfy the necessary hardware requirements. The proposed model consumes 50 kB of Flash and 12 kB of RAM and processes data locally, achieving a bandwidth reduction of 60%. Finally, the hybrid model was tested in external data sets.

      February 2023  Water Article

    Modeling and Validation of Residential Water Demand in Agent-Based Models: A Systematic Literature Review

    Bernhard Jonathan Sattler, John Friesen, Andrea Tundis, Peter F. Pelz

    PDF BibTeX DOI: 10.3390/w15030579

    Abstract
    Current challenges, such as climate change or military conflicts, show the great importance of urban supply infrastructures. In this context, an open question is how different scenarios and crises can be studied in silico to assess the interaction between the needs of social systems and technical infrastructures. Agent-based modeling is a suitable method for this purpose. This review investigates (i) how agent-based models of residential water demand should be validated, (ii) how such models are commonly built and (iii) validated, and (iv) how these validation practices compare to the recommendations drawn from question (i). Therefore, a systematic literature review using the PRISMA framework is conducted. Out of 207 screened papers, 35 models are identified with an emphasis on highly realistic models (i.e., highly detailed and representing specific real-world systems) for planning, management, and policy of urban water resources. While some models are thoroughly validated, quantified validation distinct from calibration data should be emphasized and used to communicate the confidence in results and recommendations drawn from the models. Pattern-oriented validation, validation on multiple levels and on higher moments of aggregated statistics should be considered more often. These findings expand prior literature by providing a more extensive sample of reviewed articles and recommending specific approaches for the validation of models.

      February 2023  IEEE Communications Letters Article

    Optimization-Based Phase-Shift Codebook Design for Large IRSs

    Walid R. Ghanem, Vahid Jamali, Malte Schellmann, Hanwen Cao, Joseph Eichinger, Robert Schober

    BibTeX DOI: 10.1109/LCOMM.2022.3225585

    Abstract
    In this letter, we focus on large intelligent reflecting surfaces (IRSs) and propose a new codebook construction method to obtain a set of pre-designed phase-shift configurations for the IRS unit cells. Since the complexity of online optimization and the overhead for channel estimation scale with the size of the phase-shift codebook, the design of small codebooks is of high importance. We consider both continuous and discrete phase-shift designs and formulate the codebook construction as optimization problems. To solve the optimization problems, we propose an optimal algorithm for the discrete phase-shift design and a locally optimal solution for the continuous design. Simulation results show that the proposed algorithms facilitate the construction of codebooks of different sizes and with different beamwidths. Moreover, the performance of the discrete phase-shift design with 2-bit quantization is shown to approach that of the continuous phase-shift design. Finally, our simulation results show that the proposed designs enable large transmit power savings compared to the existing linear and quadratic codebook designs.

      January 2023  2023 IEEE/CVF Winter Conference on Applications of Computer Vision Conference

    Unsupervised 4D LiDAR Moving Object Segmentation in Stationary Settings with Multivariate Occupancy Time Series

    Thomas Kreutz, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1109/WACV56688.2023.00169

    Abstract
    In this work, we address the problem of unsupervised moving object segmentation (MOS) in 4D LiDAR data recorded from a stationary sensor, where no ground truth annotations are involved. Deep learning-based state-of- the-art methods for LiDAR MOS strongly depend on anno- tated ground truth data, which is expensive to obtain and scarce in existence. To close this gap in the stationary set- ting, we propose a novel 4D LiDAR representation based on multivariate time series that relaxes the problem of un- supervised MOS to a time series clustering problem. More specifically, we propose modeling the change in occupancy of a voxel by a multivariate occupancy time series (MOTS), which captures spatio-temporal occupancy changes on the voxel level and its surrounding neighborhood. To perform unsupervised MOS, we train a neural network in a self- supervised manner to encode MOTS into voxel-level feature representations, which can be partitioned by a clustering al- gorithm into moving or stationary. Experiments on station- ary scenes from the Raw KITTI dataset show that our fully unsupervised approach achieves performance that is com- parable to that of supervised state-of-the-art approaches.

      January 2023  Elsevier Article

    Power blackout: Citizens’ contribution to strengthen local resilience

    Martin Pietsch, Michèle Knodt, Gerrit Hornung, Jan-Philipp Stroscher, Florian Steinke, Anna Stöckl

    PDF BibTeX DOI: 10.1016/j.enpol.2023.113433

    Abstract
    A long-lasting, large-scale power blackout has a huge impact on the infrastructure of public life, as well as on critical infrastructure including electricity and water supply. At the same time, it can be observed that the share of renewable energies, and thus the possibility of self-sufficiency, has increased enormously in recent years. This contribution focuses on the question to what extend citizens are willing to share their electricity resources in order to make their city more resilient. In reference to Ostrom’s concept of club or common goods, it can be shown if and how the private good of citizen’s electricity resources can be transformed into a club or even a common good. Drawing on survey data from the city of Darmstadt we investigated the willingness to share electricity and to participate in participatory formats to enhance urban resilience.

      January 2023  IEEE Transactions on Network Science and Engineering Article

    Spatial-Temporal Modeling and Analysis of Reliability and Delay in Urban V2X Networks

    Junliang Ye, Lin Xiang, Xiaohu Ge

    PDF BibTeX DOI: 10.1109/TNSE.2023.3234284

    Abstract
    The fifth-generation (5 G) based vehicle-to-everything (V2X) communication is a promising technology to enhance both manned and unmanned driving systems. To this end, the 5 G V2X networks should satisfy stringent requirements on transmission reliability and delay for exchanging safety-critical messages (SCMs). A joint analysis of transmission reliability and delay in V2X communication networks is thus crucial, particularly in urban 5 G V2X networks. This was considered prohibitive due to the complicated spatial-temporal dynamics of V2X communications caused by interference, channel fading, as well as queueing and retransmission of SCMs. Moreover, urban 5 G V2X networks are typically deployed in a finite area, where locations of nodes are spatially correlated and cannot be conveniently modeled as Poisson point process (PPP). In this paper, we propose a novel binomial point process (BPP) based analytic framework for modeling the spatial-temporal dynamics of urban 5 G V2X communications and characterizing transmission reliability and delay of SCM exchange jointly. The presented framework captures not only the spatial distribution of interference and channel fading during uplink and downlink transmissions, but also the temporal dynamics associated with queueing and retransmissions of SCMs. Exploiting the stochastic geometry theory and queueing theory, closed-form expressions of transmission reliability and delay are derived, which are further validated using Monte Carlo simulations. Both the numerical and simulation results reveal complicated couplings between the transmission reliability and delay in different operation regimes. Nevertheless, the proposed analytical framework can accurately capture the reliability-delay relations.

      2023  Sensors Article

    Energy-Efficient Decentralized Broadcasting in Wireless Multi-Hop Networks

    Artur Sterz, Robin Klose, Markus Sommer, Jonas Höchst, Jakob Link, Bernd Simon, Anja Klein, Matthias Hollick, Bernd Freisleben

    BibTeX DOI: 10.3390/s23177419

    Abstract
    Several areas of wireless networking, such as wireless sensor networks or the Internet of Things, require application data to be distributed to multiple receivers in an area beyond the transmission range of a single node. This can be achieved by using the wireless medium’s broadcast property when retransmitting data. Due to the energy constraints of typical wireless devices, a broadcasting scheme that consumes as little energy as possible is highly desirable. In this article, we present a novel multi-hop data dissemination protocol called BTP. It uses a game-theoretical model to construct a spanning tree in a decentralized manner to minimize the total energy consumption of a network by minimizing the transmission power of each node. Although BTP is based on a game-theoretical model, it neither requires information exchange between distant nodes nor time synchronization during its operation, and it inhibits graph cycles effectively. The protocol is evaluated in Matlab and NS-3 simulations and through real-world implementation on a testbed of 75 Raspberry Pis. The evaluation conducted shows that our proposed protocol can achieve a total energy reduction of up to 90% compared to a simple broadcast protocol in real-world experiments.

      2023  Technische Universität Darmstadt Darmstadt Thesis

    Decentralized Ultra-Reliable Low-Latency Communications through Concurrent Cooperative Transmission

    Robin Klose

    PDF BibTeX DOI: 10.26083/tuprints-00024070

    Abstract
    Emerging cyber-physical systems demand for communication technologies that enable seamless interactions between humans and physical objects in a shared environment. This thesis proposes decentralized URLLC (dURLLC) as a new communication paradigm that allows the nodes in a wireless multi-hop network (WMN) to disseminate data quickly, reliably and without using a centralized infrastructure. To enable the dURLLC paradigm, this thesis explores the practical feasibility of concurrent cooperative transmission (CCT) with orthogonal frequency-division multiplexing (OFDM). CCT allows for an efficient utilization of the medium by leveraging interference instead of trying to avoid collisions. CCT-based network flooding disseminates data in a WMN through a reception-triggered low-level medium access control (MAC). OFDM provides high data rates by using a large bandwidth, resulting in a short transmission duration for a given amount of data. This thesis explores CCT-based network flooding with the OFDM-based IEEE 802.11 Non-HT and HT physical layers (PHYs) to enable interactions with commercial devices. An analysis of CCT with the IEEE 802.11 Non-HT PHY investigates the combined effects of the phase offset (PO), the carrier frequency offset (CFO) and the time offset (TO) between concurrent transmitters, as well as the elapsed time. The analytical results of the decodability of a CCT are validated in simulations and in testbed experiments with Wireless Open Access Research Platform (WARP) v3 software-defined radios (SDRs). CCT with coherent interference (CI) is the primary approach of this thesis. Two prototypes for CCT with CI are presented that feature mechanisms for precise synchronization in time and frequency. One prototype is based on the WARP v3 and its IEEE 802.11 reference design, whereas the other prototype is created through firmware modifications of the Asus RT-AC86U wireless router. Both prototypes are employed in testbed experiments in which two groups of nodes generate successive CCTs in a ping-pong fashion to emulate flooding processes with a very large number of hops. The nodes stay synchronized in experiments with 10 000 successive CCTs for various modulation and coding scheme (MCS) indices and MAC service data unit (MSDU) sizes. The URLLC requirement of delivering a 32-byte MSDU with a reliability of 99.999 % and with a latency of 1 ms is assessed in experiments with 1 000 000 CCTs, while the reliability is approximated by means of the frame reception rate (FRR). An FRR of at least 99.999 % is achieved at PHY data rates of up to 48 Mbit/s under line-of-sight (LOS) conditions and at PHY data rates of up to 12 Mbit/s under non-line-of-sight (NLOS) conditions on a 20 MHz wide channel, while the latency per hop is 48.2 µs and 80.2 µs, respectively. With four multiple input multiple output (MIMO) spatial streams on a 40 MHz wide channel, a LOS receiver achieves an FRR of 99.5 % at a PHY data rate of 324 Mbit/s. For CCT with incoherent interference, this thesis proposes equalization with time-variant zero-forcing (TVZF) and presents a TVZF receiver for the IEEE 802.11 Non-HT PHY, achieving an FRR of up to 92 % for CCTs from three unsyntonized commercial devices. As CCT-based network flooding allows for an implicit time synchronization of all nodes, a reception-triggered low-level MAC and a reservation-based high-level MAC may in combination support various applications and scenarios under the dURLLC paradigm.

      January 2023  International Journal of Machine Learning and Cybernetics Article

    Data augmentation in natural language processing: a novel text generation approach for long and short text classifiers

    Markus Bayer, Marc-André Kaufhold, Björn Buchhold, Marcel Keller, Jörg Dallmeyer, Christian Reuter

    BibTeX DOI: 10.1007/s13042-022-01553-3

    Abstract
    In many cases of machine learning, research suggests that the development of training data might have a higher relevance than the choice and modelling of classifiers themselves. Thus, data augmentation methods have been developed to improve classifiers by artificially created training data. In NLP, there is the challenge of establishing universal rules for text transformations which provide new linguistic patterns. In this paper, we present and evaluate a text generation method suitable to increase the performance of classifiers for long and short texts. We achieved promising improvements when evaluating short as well as long text tasks with the enhancement by our text generation method. Especially with regard to small data analytics, additive accuracy gains of up to 15.53% and 3.56% are achieved within a constructed low data regime, compared to the no augmentation baseline and another data augmentation technique. As the current track of these constructed regimes is not universally applicable, we also show major improvements in several real world low data tasks (up to +4.84 F1-score). Since we are evaluating the method from many perspectives (in total 11 datasets), we also observe situations where the method might not be suitable. We discuss implications and patterns for the successful application of our approach on different types of datasets.

      2023  Designing Resilience Global (DRG) International Symposium and Competition Conference

    Designing Resilience Global (DRG) - International Symposium and Competition - Documentation

    PDF BibTeX DOI: 10.26083/tuprints-00023774

    Abstract
    The 2022 Designing Resilience Global (DRG) Symposium and Competition took part from June 20th to June 24th. The documentation includes transcripts of all keynote speeches of renown experts in academia and practice and the submissions of all universities that took part in the international competition.

      2023  Mobility Design: Die Zukunft der Mobilität gestalten Band 2: Forschung Other

    Mobilität als Schlüssel zur lebenswerten Stadt

    Björn Hekmati, Annette Rudolph-Cleff

    PDF BibTeX DOI: 10.1515/9783868597936

    Abstract
    Klimawandel und Ressourcenverknappung, aber auch der stetig steigende Verkehrsaufwand machen es unabdingbar, neue Lösungen für eine umweltschonende und menschenfreundliche Mobilität zu entwickeln. Mit dem Ausbau digitaler Informationssysteme werden wir zukünftig unterschiedliche Verkehrsträger entsprechend unseren Bedürfnissen leicht kombinieren können. Diese Entwicklungen sind für die Gestaltung verschiedener Mobilitätsräume eine große Herausforderung. Lag der Schwerpunkt in Band 1 auf der Praxis, versammelt Band 2 nun Forschungen aus den Bereichen Design, Architektur, Stadtplanung, Geografie, Sozialwissenschaft, Verkehrsplanung, Psychologie und Kommunikationstechnologie. Die aktuelle Diskussion über die Verkehrswende wird um die Perspektive des nutzer*innenzentrierten Mobilitätsdesigns erweitert.

      2023  26th International Conference on Artificial Intelligence and Statistics Conference

    Learning Sparse Graphon Mean Field Games

    Christian Fabian, Kai Cui, Heinz Koeppl

    PDF BibTeX

    Abstract
    Although the field of multi-agent reinforcement learning (MARL) has made considerable progress in the last years, solving systems with a large number of agents remains a hard challenge. Graphon mean field games (GMFGs) enable the scalable analysis of MARL problems that are otherwise intractable. By the mathematical structure of graphons, this approach is limited to dense graphs which are insufficient to describe many real-world networks such as power law graphs. Our paper introduces a novel formulation of GMFGs, called LPGMFGs, which leverages the graph theoretical concept of Lp graphons and provides a machine learning tool to efficiently and accurately approximate solutions for sparse network problems. This especially includes power law networks which are empirically observed in various application areas and cannot be captured by standard graphons. We derive theoretical existence and convergence guarantees and give empirical examples that demonstrate the accuracy of our learning approach for systems with many agents. Furthermore, we extend the Online Mirror Descent (OMD) learning algorithm to our setup to accelerate learning speed, empirically show its capabilities, and conduct a theoretical analysis using the novel concept of smoothed step graphons. In general, we provide a scalable, mathematically well-founded machine learning approach to a large class of otherwise intractable problems of great relevance in numerous research fields.

      2023  INFORMATIK 2023 - 53. Jahrestagung der Gesellschaft für Informatik e.V. Conference

    Das Netz hat Geschichte

    Jonas Franken, Marco Zivkovic, Nadja Thiessen, Jens Ivo Engels, Christian Reuter

    BibTeX DOI: 10.18420/inf2023_159

    Abstract
    Kritische Infrastrukturen sind häufig über Jahrzehnte gewachsene, komplexe Netze. Den- noch fehlt derzeit die historische Perspektive auf die Aufschichtungstendenzen von Technologien in den Sektoren, die für die Gesellschaft essenzielle Dienste bereitstellen. Ein besseres Verständnis von Ausbreitungs-, Ausbau-, Ersatz- und Ausmusterungsprozessen kann Entscheidungshilfe und Orientierung für resilientere Versorgungsnetzarchitekturen in der Zukunft geben. Kompatibilitäts- probleme mit Legacy-Soft- und Hardware sind bekannte Phänomene in vielen KRITIS-Einrichtun- gen. Entsprechend gewinnen Wissens- und Erfahrungstransfers bei zunehmend komplexen, dennoch über Jahrzehnte verwendete Technologien in landwirtschaftlichen Betrieben enorm an Bedeutung. Der Beitrag vollzieht die Konzeption und Fragestellungen eines interdisziplinären Forschungspro- jekts nach, in welchem die Verwundbarkeit der kritischen Infrastruktursektoren Verkehr und Kom- munikation im Rhein-Main-Gebiet analysiert wird. Von den Leistungen beider Sektoren hängt die digitale Landwirtschaft stark ab. Insbesondere rurale, beim digitalen und Schienennetzausbau häufig vernachlässigte Gebiete werden dabei mittels explorativer Interviewstudie und anschließender ar- chivbasierter, quantitativer Überprüfung der zuvor generierten Hypothesen aus einer raum-zeitli- chen und technischen Perspektive untersucht.

      2023  WiSec ‘23: 16th ACM Conference on Security and Privacy in Wireless and Mobile Networks Conference

    DEMO: Secure Bootstrapping of Smart Speakers Using Acoustic Communication

    Markus Scheck, Florentin Putz, Frank Hessel, Hermann Leinweber, Jonatan Crystall, Matthias Hollick

    PDF BibTeX DOI: 10.26083/tuprints-00024180

    Abstract
    Smart speakers are highly privacy-sensitive devices: They are located in our homes and provide an Internet-enabled microphone, making them a prime target for attackers. The pairing between a client device and the speaker must be protected to prohibit adversaries from accessing the device. Most commercial protocols are vulnerable to nearby adversaries as they do not probe for human presence at the speaker or proximity between both devices. In addition to security, the protocol must provide a user-friendly way for initial bootstrapping of the speaker. We design an open pairing protocol for the establishment of a shared secret between both devices using acoustic messaging to guarantee proximity, and release our implementation for the smart speaker as well as Android and Linux clients as open-source software on GitHub.

      2023  European Societies Article

    Civil society and sense of community in Ukraine: from dormancy to action

    Kateryna Zarembo, Eric Martin

    BibTeX DOI: 10.1080/14616696.2023.2185652

    Abstract
    The academic literature offers different views on the strength of Ukraine’s civil society, but Ukraine’s massive civic engagement and collective action, most recently in defense against Russian aggression, offers a startling picture of grass-root activism. Based on interviews, surveys and archival research, we highlight changes and nuances to Ukrainian civil society, civic engagement and motivations over time, from Euromaidan, through the hybrid Russian aggression in the East, to the recent full-scale Russian invasion. In doing so, we explore a more inclusive understanding of civil society complemented by sense of community and community responsibility.

      2023  Technische Universität Darmstadt Berlin Thesis

    Gefährdung städtischer Infrastruktur durch Hochwasser: Wahrnehmungen und Bewältigungsstrategien in Mannheim und Dresden 1918-1989

    Nadja Thiessen

    BibTeX DOI: 10.1515/9783110734676

    Abstract
    Mannheim und Dresden sind seit ihrer Gründung durch die Dynamiken des Wassers geprägt. Vor allem die zyklisch auftretenden Hochwasser führen zu Gefährdungen. Im kurzen 20. Jahrhundert existierte ein konstanter Umgang damit innerhalb der Städte. Die Studie fokussiert diese Bewältigungsstrategien, die unter dem Einfluss lokalspezifischer Faktoren wie der Stadtentwicklung sowie übergeordneten politischen und wirtschaftlichen Bedingungen standen.

      2023  Technische Universität Darmstadt Darmstadt Thesis

    Aerial Network Assistance Systems for Post-Disaster Scenarios : Topology Monitoring and Communication Support in Infrastructure-Independent Networks

    Julian Zobel

    PDF BibTeX DOI: 10.26083/tuprints-00023043

    Abstract
    Communication anytime and anywhere is necessary for our modern society to function. However, the critical network infrastructure quickly fails in the face of a disaster and leaves the affected population without means of communication. This lack can be overcome by smartphone-based emergency communication systems, based on infrastructure-independent networks like Delay-Tolerant Networks (DTNs). DTNs, however, suffer from short device-to-device link distances and, thus, require multi-hop routing or data ferries between disjunct parts of the network. In disaster scenarios, this fragmentation is particularly severe because of the highly clustered human mobility behavior. Nevertheless, aerial communication support systems can connect local network clusters by utilizing Unmanned Aerial Vehicles (UAVs) as data ferries. To facilitate situation-aware and adaptive communication support, knowledge of the network topology, the identification of missing communication links, and the constant reassessment of dynamic disasters are required. These requirements are usually neglected, despite existing approaches to aerial monitoring systems capable of detecting devices and networks. In this dissertation, we, therefore, facilitate the coexistence of aerial topology monitoring and communications support mechanisms in an autonomous Aerial Network Assistance System for infrastructure-independent networks as our first contribution. To enable system adaptations to unknown and dynamic disaster situations, our second contribution addresses the collection, processing, and utilization of topology information. For one thing, we introduce cooperative monitoring approaches to include the DTN in the monitoring process. Furthermore, we apply novel approaches for data aggregation and network cluster estimation to facilitate the continuous assessment of topology information and an appropriate system adaptation. Based on this, we introduce an adaptive topology-aware routing approach to reroute UAVs and increase the coverage of disconnected nodes outside clusters. We generalize our contributions by integrating them into a simulation framework, creating an evaluation platform for autonomous aerial systems as our third contribution. We further increase the expressiveness of our aerial system evaluation, by adding movement models for multicopter aircraft combined with power consumption models based on real-world measurements. Additionally, we improve the disaster simulation by generalizing civilian disaster mobility based on a real-world field test. With a prototypical system implementation, we extensively evaluate our contributions and show the significant benefits of cooperative monitoring and topology-aware routing, respectively. We highlight the importance of continuous and integrated topology monitoring for aerial communications support and demonstrate its necessity for an adaptive and long-term disaster deployment. In conclusion, the contributions of this dissertation enable the usage of autonomous Aerial Network Assistance Systems and their adaptability in dynamic disaster scenarios.

      2023  Universitäts- und Landesbibliothek Darmstadt Darmstadt Other

    Urbane Datenplattformen und Resilienz der Städte: Status quo in Deutschland und Empfehlungen für kommunale Akteure

    Michaela Leštáková, Lucía Wright-Contreras, Leonie Schiermeyer

    PDF BibTeX DOI: 10.26083/tuprints-00024392

    Abstract
    Digitalisierung in Städten ist ein globaler Trend. Auch in der Bundesrepublik Deutschland streben viele Städte an, sogenannte Smart Cities zu werden. Um dieses Ziel zu erreichen, werden in Städten zunehmend Daten zu verschiedenen Themenbereichen gesammelt – mit der Absicht, den Informationsaustausch zwischen Behörden, Bürgerinnen und Bürgern und anderen städtischen Akteuren einfacher, effizienter und transparenter zu gestalten. Zu diesem Zweck werden häufig sogenannte urbane Datenplattformen eingesetzt. In diesem Praxisdossier wird die Rolle von urbanen Datenplattformen im Kontext von Resilienz und Smart Cities untersucht. In Kooperation zwischen dem LOEWE-Zentrum emergenCITY und Haselhorst Associates wurden zu diesem Zweck zwei Forschungsfragen definiert: 1) Können urbane Datenplattformen dazu beitragen, die Resilienz der Stadt zu erhöhen? (Resilienz durch IKT) und 2) Sind die existierenden urbanen Datenplattformen in sich resilient? (Resilienz für IKT). Zur Beantwortung dieser Fragen wurden über 400 deutsche Städte hinsichtlich des Vorhandenseins oder der geplanten Einführung einer städtischen Datenplattform untersucht. Ergänzt wurden die Ergebnisse durch eine Analyse der Smart-Cities-Modellprojekte. Basierend auf der Analyse werden 6 Empfehlungen für kommunale Akteure formuliert.

      2023  International Conference on Modelling and Applied Simulation (MAS 2023) Conference

    A framework for the simulation-based selection of social models for socio-technical models of infrastructures using technical requirements analysis

    Bernhard Jonathan Sattler, Jannik Stadler, Andrea Tundis, John Friesen, Peter F. Pelz

    PDF BibTeX DOI: 10.46354/i3m.2023.mas.010

    Abstract
    Urbanization increases the importance of urban infrastructures, with computer models and simulation being important tools for their planning and management. Human factors are increasingly included into infrastructure models, creating socio-technical models. This paper proposes a novel framework for selecting these social (sub-)models. For this, requirements analysis of the technical system is used to identify critical physical parameters. The impact of different assumptions in the social model on the critical physical parameters are determined using simulation and hypothesis testing. This impact is used to determine the relevance of the differing assumptions and to select the right social model. Finally, a preliminary case study of the water distribution system of Darmstadt, Germany, is used to show the efficacy of the framework by comparing two water demand models. The results of the case study show, that the framework can be used to quantify the relevant system behavior and test the significance of model assumptions.

      2023  Environment Systems and Decisions Article

    Resilience beyond insurance: coordination in crisis governance

    Eva Katharina Platzer, Michèle Knodt

    PDF BibTeX DOI: 10.1007/s10669-023-09938-7

    Abstract
    The latest report by the Intergovernmental Panel on Climate Change (IPCC) warns of an increase in heavy rainfall events due to global warming and climate change, which can result in significant economic costs for insurance companies and businesses. To address this challenge, insurance companies are focusing on developing new risk management strategies and offering new products such as flood insurance. However, the article argues that effective and feasible coordination shortens recovery time and can therefore drastically reduce the financial costs of a crisis—that is, the insurance costs. The paper analyses the deficit in crisis management during heavy rain events in Germany, based on the 2021 Ahr valley flood. The analysis is conducted based on document analysis and interviews and focuses on three areas of deficit: coordination between crisis staffs and (1) civil society, (2) emergency responders, and (3) political leaders. The paper highlights the importance of coordination during a crisis, which can help to address the crisis more efficiently and effectively, minimise damage and get communities back on their feet faster. The paper recommends policy changes to improve interface management and disaster management coordination.

      December 2022  Engineering Proceedings Article

    Analysis of Helicopter Flights in Urban Environments for UAV Traffic Management

    David Hünemohr, Maximilian Bauer, Jan Kleikemper, Markus Peukert

    PDF BibTeX DOI: 10.3390/engproc2022028010

    Abstract
    Future air mobility will consist of increased unmanned aerial vehicle (UAV) traffic operating in urban areas. Currently, the lower airspace in these environments is mainly used by traffic operating under visual flight rules, particularly helicopters in emergency medical services (HEMS). In the presented work, we analyze urban HEMS missions with automatic dependent surveillance-broadcast (ADS-B) data to identify the potential benefits to support UAV traffic management (UTM). In our methodology, we first restrict an existing HEMS ADS-B data set to a specific city and then further process it to extract the valid HEMS flights. Because no other mission information is available, we apply rule-based algorithms to define different helicopter flight segments and characterize specific HEMS mission segments. The resulting data set is analyzed to extract the characteristic information about the HEMS traffic within the city. The methodology is applied to the ADS-B HEMS flight data in the area of Berlin. The results show that the HEMS and flight segments can be identified robustly, and specific flight patterns are characteristic for them. Based on the results of this analysis, UAV traffic alert strategies are proposed to demonstrate the potential benefit of integrating ADS-B data statistics for UTM.

      December 2022  33rd International Symposium on Personal, Indoor and Mobile Radio Communications Conference

    Connectivity Analysis for Large-Scale Intelligent Reflecting Surface Aided mmWave Cellular Networks

    Yi Wang, Lin Xiang, Jing Zhang, Xiaohu Ge

    BibTeX DOI: 10.1109/PIMRC54779.2022.9977979

    Abstract
    This paper presents a stochastic geometry framework for modeling and evaluating the connectivity of uplink transmission in a large-scale intelligent reflecting surface (IRS) assisted millimeter-wave (mmWave) communication network, where the uplink user equipments (UEs) attempt to communicate with the nearest base stations (BSs) either without or with the help of an IRS. We propose a novel elliptical geometry model, which can effectively capture the impact of IRS location and orientation, as well as incident/reflection angle on mmWave signal propagation, while, at the same time, significantly simplifying the analysis of the system performance. Employing the elliptical geometry model, the approximate reflection probability of IRS as well as its upper and lower bounds are derived in closed form. Based on these results, we further analyze the successful connection probability of uplink UEs for IRS-assisted mmWave cellular networks. Our results show that compared with conventional direct UE-to-BS communication without IRS, indirect communication with the aid of IRS exhibits a slower decaying in the connection probability as the communication distance increases, as the latter can significantly increase the connection probability for cell-edge UEs. Moreover, for mmWave BSs with small receiving power thresholds, the deployment of IRS can effectively mitigate the impact of blockages to improve mmWave signal propagation.

      December 2022  33rd International Symposium on Personal, Indoor and Mobile Radio Communications Conference

    UAV-Assisted Delay-Sensitive Communications with Uncertain User Locations: A Cost Minimization Approach

    Burak Yilmaz, Lin Xiang, Anja Klein

    BibTeX DOI: 10.1109/PIMRC54779.2022.9977912

    Abstract
    In this paper, we consider optimal resource allocation for unmanned aerial vehicle (UAV)-assisted delay-sensitive communications, where a UAV flies to deliver time-critical messages to multiple ground users (GUs) as soon as possible. However, the GUs’ locations cannot be perfectly known at the UAV, which may jeopardize the timeliness of message delivery to the GUs. To tackle this challenge, we consider a disk-based fixed-rate transmission scheme at the UAV, which can exploit the mobility of the UAV to facilitate timely communications despite uncertain user locations. Consequently, the system performance hinges on the UAV’s flight trajectory and the scheduling of GUs, which are further optimized using a cost minimization approach. Thereby, a general class of delay-aware cost functions, referred to as the cost of delivery delay (CoDD), is defined taking into account the diverse delay-sensitivity requirements of the GUs, and we jointly optimize the user scheduling and the UAV’s trajectory for minimization of the sum CoDD of all GUs incurred before the UAV’s mission completes. The formulated optimization problem is a nonconvex mixed-integer nonlinear program. Exploiting the underlying structure of this problem, we further propose two novel low-complexity solutions based on approximate dynamic programming (DP). Simulation results show that the proposed schemes can flexibly adjust the UAV’s flight trajectory and resource allocation according to the GUs’ individual delivery delays, delay tolerance, and location uncertainty, which translates into significantly lower sum CoDD for the GUs than several benchmark schemes.

      December 2022  2022 IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR) Conference

    Visualization of Machine Learning Uncertainty in AR-Based See-Through Applications

    Achref Doula, Lennart Schmidt, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1109/AIVR56993.2022.00022

    Abstract
    Augmented reality see-through applications rely mostly on machine learning models to detect and localize occluded objects. In this case, the user is usually presented the result with the highest probability without taking into account the uncertainty of the model. However, the uncertainty plays a vital role when considering applications where a critical decision-making process relies heavily on the predictions of the model, such as in the case where occluded cars are shown to a driver. In this work, we conduct an investigation of the effects of communicating the uncertainty of machine learning models to users in AR-based see-through applications. Through a controlled user study, we compare three visualization modes: no visualization, most probable output, and probability distribution. The results of our evaluation reveal that when considering the visualizations, each of them lead to comparable results in terms of speed and accuracy of the decision-making process. A relevant finding is that participants considered uncertainty as a substantial part of the output of machine learning models and needs to be delivered with the results. An additional important conclusion is that the preference of users over a specific visualization is strongly dependent on the particular use case.

      December 2022  ACM Computing Surveys Article

    Survey on Data Augmentation for Text Classification

    Markus Bayer, Marc-André Kaufhold, Christian Reuter

    PDF BibTeX DOI: 10.1145/3544558

    Abstract
    Data augmentation, the artificial creation of training data for machine learning by transformations, is a widely studied research field across machine learning disciplines. While it is useful for increasing a model’s generalization capabilities, it can also address many other challenges and problems, from overcoming a limited amount of training data, to regularizing the objective, to limiting the amount data used to protect privacy. Based on a precise description of the goals and applications of data augmentation and a taxonomy for existing works, this survey is concerned with data augmentation methods for textual classification and aims to provide a concise and comprehensive overview for researchers and practitioners. Derived from the taxonomy, we divide more than 100 methods into 12 different groupings and give state-of-the-art references expounding which methods are highly promising by relating them to each other. Finally, research perspectives that may constitute a building block for future work are provided.

      December 2022  Die Friedens-Warte : journal of international peace and organization Article

    Trustworthy and Explainable: A European Vision of (Weaponised) Artificial Intelligence

    Stefka Schmid

    BibTeX DOI: 10.35998/fw-2022-0013

    Abstract
    The European Union (EU) has undertaken policies which address the research and development of artificial intelligence (AI). In light of debates of technology assessment which focus on risks for humans and questions of control of AI, the EU has propagated an ethical, human-centred approach of the application of AI. It is important to identify how the EU envisions AI as this may guide emerging norms in AI governance and today’s research and development of (weaponised) AI. Building on works of Human-Computer Interaction (HCI), this work derives the actor’s understanding of human-AI inter- action, including conceptualisations of explainability, interpretability, and risks. Analysis of EU docu- ments on the implementation of AI as a general-purpose technology and for military application reveals that explainability and risk identification are crucial elements for trust, which itself is a necessary component in the uptake of AI. Interdisciplinary approaches allow for a more detailed understanding of actors’ fundamental views on human control of AI, which further contributes to debates on technology assessment in professionalised political contexts.

      December 2022  GLOBECOM 2022 - 2022 IEEE Global Communications Conference Conference

    Deep Reinforcement Learning for Task Allocation in Energy Harvesting Mobile Crowdsensing

    Sumedh Dongare, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX DOI: 10.1109/GLOBECOM48099.2022.10001204

    Abstract
    Mobile crowd-sensing (MCS) is an upcoming sensing architecture which provides better coverage, accuracy, and requires lower costs than traditional wireless sensor networks. It utilizes a collection of sensors, or crowd, to perform various sensing tasks. As the sensors are battery operated and require a mechanism to recharge them, we consider energy harvesting (EH) sensors to form a sustainable sensing architecture. The execution of the sensing tasks is controlled by the mobile crowd-sensing platform (MCSP) which makes task allocation decisions, i.e., it decides whether or not to perform a task depending on the available resources, and if the task is to be performed, assigns it to suitable sensors. To make optimal allocation decisions, the MCSP requires perfect non-causal knowledge regarding the channel coefficients of the wireless links to the sensors, the amounts of energy the sensors harvest and the sensing tasks to be performed. However, in practical scenarios this non-causal knowledge is not available at the MCSP. To overcome this problem, we propose a novel Deep-Q-Network solution to find the task allocation strategy that maximizes the number of completed tasks using only realistic causal knowledge of the battery statuses of the available sensors. Through numerical evaluations we show that our proposed approach performs only 7.8% lower than the optimal solution. Moreover, it outperforms the myopically optimal and the random task allocation schemes.

      December 2022  GLOBECOM 2022 - 2022 IEEE Global Communications Conference Conference

    RIS-assisted beamforming for energy efficiency in multiuser downlink transmissions

    Jaime Quispe, Tarcisio Ferreira Maciel, Yuri Carvalho Barbosa Silva, Anja Klein

    BibTeX DOI: 10.1109/GCWkshps56602.2022.10008573

    Abstract
    Reconfigurable intelligent surface (RIS) based reflections is a promising approach to increase spectral efficiency (SE) and reduce power consumption of wireless communications systems. This paper investigates the trade-off between these two metrics by considering the energy efficiency (EE) maximization of an RIS-assisted multiuser downlink transmission from a multiantenna base station (BS) to multiple single-antenna users while satisfying constraints on quality-of-service (QoS), RIS phase shifts, and BS maximum transmit power. We consider a coordinated beamforming scheme and propose a joint optimization procedure based on the Dinkelbach, fractional programming, and semi-definite relaxation (SDR) methods. Simulation results show that the RIS-assisted system is more energy-efficient than its counterpart without RIS and that the RIS, particularly when it is equipped with a large number of antenna elements, can simultaneously improve the SE and power consumption of the transmission. Furthermore, the presented algorithm achieves good-quality solutions that are competitive to the obtained via exhaustive search with branch-reduce-and-bound (BRnB) methods and requires fewer iterations to converge.

      December 2022 Other

    CySecBERT: A Domain-Adapted Language Model for the Cybersecurity Domain

    Markus Bayer, Philipp Kuehn, Ramin Shanehsaz, Christian Reuter

    PDF BibTeX DOI: 10.48550/arXiv.2212.02974

    Abstract
    The field of cybersecurity is evolving fast. Experts need to be informed about past, current and - in the best case - upcoming threats, because attacks are becoming more advanced, targets bigger and systems more complex. As this cannot be addressed manually, cybersecurity experts need to rely on machine learning techniques. In the texutual domain, pre-trained language models like BERT have shown to be helpful, by providing a good baseline for further fine-tuning. However, due to the domain-knowledge and many technical terms in cybersecurity general language models might miss the gist of textual information, hence doing more harm than good. For this reason, we create a high-quality dataset and present a language model specifically tailored to the cybersecurity domain, which can serve as a basic building block for cybersecurity systems that deal with natural language. The model is compared with other models based on 15 different domain-dependent extrinsic and intrinsic tasks as well as general tasks from the SuperGLUE benchmark. On the one hand, the results of the intrinsic tasks show that our model improves the internal representation space of words compared to the other models. On the other hand, the extrinsic, domain-dependent tasks, consisting of sequence tagging and classification, show that the model is best in specific application scenarios, in contrast to the others. Furthermore, we show that our approach against catastrophic forgetting works, as the model is able to retrieve the previously trained domain-independent knowledge. The used dataset and trained model are made publicly available.

      December 2022  38th Annual Computer Security Applications Conference Conference

    Ripples in the Pond: Transmitting Information through Grid Frequency Modulation

    Jan Sebastian Götte, Liran Katzir, Björn Scheuermann

    PDF BibTeX DOI: 10.1145/3564625.3564640

    Abstract
    The growing heterogenous ecosystem of networked consumer devices such as smart meters or IoT-connected appliances such as air conditioners is difficult to secure, unlike the utility side of the grid which can be defended effectively through rigorous IT security measures such as isolated control networks. In this paper, we consider a crisis scenario in which an attacker compromises a large number of consumer-side devices and modulates their electrical power to destabilize the grid and cause an electrical outage [9, 26, 27, 47, 50, 55]. In this paper propose a broadcast channel based on the modulation of grid frequency through which utility operators can issue commands to devices at the consumer premises both during an attack for mitigation and in its wake to aid recovery. Our proposed grid frequency modulation (GFM) channel is independent of other telecommunication networks. It is resilient towards localized blackouts and it is operational immediately after power is restored. Based on our GFM broadcast channel we propose a “safety reset” system to mitigate an ongoing attack by disabling a device’s network interfaces and resetting its control functions. It can also be used in the wake of an attack to aid recovery by shutting down non-essential loads to reduce strain on the grid. To validate our proposed design, we conducted simulations based on measured grid frequency behavior. Based on these simulations, we performed an experimental validation on simulated grid voltage waveforms using a smart meter equipped with a prototype safety reset system based on a commodity microcontroller.

      December 2022  2022 Conference on Empirical Methods in Natural Language Processing Conference

    The challenges of temporal alignment on Twitter during crisis

    Aniket Pramanick, Tilman Beck, Kevin Stowe, Iryna Gurevych

    PDF BibTeX

    Abstract
    Language use changes over time, and this impacts the effectiveness of NLP systems. This phenomenon is even more prevalent in social media data during crisis events where meaning and frequency of word usage may change over the course of days. Contextual language models fail to adapt temporally, emphasizing the need for temporal adaptation in models which need to be deployed over an extended period of time. While existing approaches consider data spanning large periods of time (from years to decades), shorter time spans are critical for crisis data. We quantify temporal degradation for this scenario and propose methods to cope with performance loss by leveraging techniques from domain adaptation. To the best of our knowledge, this is the first effort to explore effects of rapid language change driven by adversarial adaptations, particularly during natural and human-induced disasters. Through extensive experimentation on diverse crisis datasets, we analyze under what conditions our approaches outperform strong baselines while highlighting the current limitations of temporal adaptation methods in scenarios where access to unlabeled data is scarce.

      November 2022  IEEE PES Innovative Smart Grid Technology (ISGT Europe 2022) Conference

    Monitoring Electricity Demand Synchronization Using Copulas

    Tobias Gebhard, Florian Steinke, Eva Brucherseifer

    BibTeX DOI: 10.1109/ISGT-Europe54678.2022.9960369

    Abstract
    Synchronization of the behavior of residential consumers, for example during crises, can lead to overloads in electric power grids. This holds especially for distribution grids, where the electrical infrastructure is not designed for the simultaneous high consumption of all households. Therefore, the monitoring and detection of (upcoming) synchronization trends is important. It is the basis for any countermeasures. We propose to model the dependency structure of consumer demands with a Gaussian copula using its correlation parameter as an indicator for synchronization. We then analyze the probability distribution of the aggregated load depending on the synchronization indicator. This allows us to infer the synchronization parameter from load measurements in real-time using a Bayesian approach. In simulation experiments with realistic household consumption distributions, we show how increased synchronization can be detected.

      November 2022  IEEE PES Innovative Smart Grid Technology (ISGT Europe 2022) Conference

    Optimized UAV Placement for Resilient Crisis Communication and Power Grid Restoration

    Michael Heise, Martin Pietsch, Florian Steinke, Maximilian Bauer, Burak Yilmaz

    BibTeX DOI: 10.1109/ISGT-Europe54678.2022.9960494

    Abstract
    During crises where both communication networks and the electricity grid break down, restoring each individual infrastructure for disaster relief becomes generally infeasible. To tackle this challenge, we propose a disaster management solution using mobile ad-hoc networks (MANETs) formed by unmanned aerial vehicles (UAVs), offering a promising solution for emergency response. Apart from establishing emergency communications for rescue teams, UAV-enabled MANETs can also enable the formation of electrical microgrids based on distributed energy resources (DER) to locally restore the electric power. We determine the optimal locations and the number of UAVs for this purpose, taking the UAVs’ needs for repeated recharging into account. The problem is formulated on a discrete grid of potential places as a mixed-integer linear program (MILP) and solved via an accelerated feasibility query algorithm (FQA). The framework is evaluated for a toy-example and a modified version of the IEEE 123 node test feeder. Simulation results show that compared with conventional MILP approaches, the proposed FQA algorithm can significantly lower the computation times, particularly for large scale MANETs.

      November 2022 Book

    A European Perspective on Crisis Informatics: Citizens‘ and Authorities‘ Attitudes Towards Social Media for Public Safety and Security

    Christian Reuter

    PDF BibTeX DOI: 10.1007/978-3-658-39720-3

    Abstract
    Mobilising helpers in the event of a flood or letting friends know that you are okay in the event of a terrorist attack – more and more people are using social media in emergency, crisis or disaster situations. Storms, floods, attacks or pandemics (esp. COVID-19) show that citizens use social media to inform themselves or to coordinate. This book presents qualitative and quantitative studies on the attitudes of emergency services and citizens in Europe towards social media in emergencies. Across the individual sub-studies, almost 10,000 people are surveyed including representative studies in the Netherlands, Germany, the UK and Italy. The work empirically shows that social media is increasingly important for emergency services, both for prevention and during crises; that private use of social media is a driving force in shaping opinions for organisational use; and that citizens have high expectations towards authorities, especially monitoring social media is expected, and sometimes responses within one hour. Depending on the risk culture, the data show further differences, e.g. whether the state (Germany) or the individual (Netherlands) is seen as primarily responsible for coping with the situation.

      November 2022  25th IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR 2021) Conference

    3D Coverage Path Planning for Efficient Construction Progress Monitoring

    Katrin Becker, Martin Oehler, Oskar von Stryk

    BibTeX DOI: 10.1109/SSRR56537.2022.10018726

    Abstract
    On construction sites, progress must be monitored continuously to ensure that the current state corresponds to the planned state in order to increase efficiency, safety and detect construction defects at an early stage. Autonomous mobile robots can document the state of construction with high data quality and consistency. However, finding a path that fully covers the construction site is a challenging task as it can be large, slowly changing over time, and contain dynamic objects. Existing approaches are either exploration approaches that require a long time to explore the entire building, object scanning approaches that are not suitable for large and complex buildings, or planning approaches that only consider 2D coverage. In this paper, we present a novel approach for planning an efficient 3D path for progress monitoring on large construction sites with multiple levels. By making use of an existing 3D model we ensure that all surfaces of the building are covered by the sensor payload such as a 360-degree camera or a lidar. This enables the consistent and reliable monitoring of construction site progress with an autonomous ground robot. We demonstrate the effectiveness of the proposed planner on an artificial and a real building model, showing that much shorter paths and better coverage are achieved than with a traditional exploration planner.

      November 2022  Chaos: An Interdisciplinary Journal of Nonlinear Science Article

    Hypergraphon mean field games

    Kai Cui, Wasiur R. KhudaBukhsh, Heinz Koeppl

    BibTeX DOI: 10.1063/5.0093758

    Abstract
    We propose an approach to modeling large-scale multi-agent dynamical systems allowing interactions among more than just pairs of agents using the theory of mean field games and the notion of hypergraphons, which are obtained as limits of large hypergraphs. To the best of our knowledge, ours is the first work on mean field games on hypergraphs. Together with an extension to a multi-layer setup, we obtain limiting descriptions for large systems of non-linear, weakly interacting dynamical agents. On the theoretical side, we prove the well-foundedness of the resulting hypergraphon mean field game, showing both existence and approximate Nash properties. On the applied side, we extend numerical and learning algorithms to compute the hypergraphon mean field equilibria. To verify our approach empirically, we consider a social rumor spreading model, where we give agents intrinsic motivation to spread rumors to unaware agents, and an epidemic control problem. Recent developments in the field of complex systems have shown that real-world multi-agent systems are often not restricted to pairwise interactions, bringing to light the need for tractable models allowing higher-order interactions. At the same time, the complexity of analysis of large-scale multi-agent systems on graphs remains an issue even without considering higher-order interactions. An increasingly popular and tractable approach of analysis is the theory of mean field games. We combine mean field games with higher-order structure by means of hypergraphons, a limiting description of very large hypergraphs. To motivate our model, we build a theoretical foundation for the limiting system, showing that the limiting system has a solution and that it approximates finite, sufficiently large systems well. This allows us to analyze otherwise intractable, large hypergraph games with theoretical guarantees, which we verify using two examples of rumor spreading and epidemics control.

      November 2022  Electric Power Systems Research Article

    The water energy nexus: Improved emergency grid restoration with DERs

    Martin Pietsch, Florian Steinke

    BibTeX DOI: 10.1016/j.epsr.2022.108468

    Abstract
    Water networks as critical infrastructures typically feature emergency electricity generators for bridging short power blackouts. We propose to combine these black start capable generators with available distributed energy resources (DERs) in the power grid, often photovoltaic generation, to jointly restore both the electricity and the water grid in the case of emergencies. This is mutually beneficial for both notworks since common grid-following inverters of DERs cannot supply power without a grid-forming nucleus. We model both grids as a coupled graph and formulate a stochastic mixed-integer linear program to determine optimal switch placement and/or optimal switching sequences jointly for both networks. Limited fuel and power availabilities, grid-forming constraints, storages, and an even distribution of available resources are considered. By minimizing the number of switching devices and switching events we target manual operability. The proposed method extends the time that can be bridged until a full restoration of the main power grid is achieved. For a small example, we demonstrate that given enough solar radiation our solution allows us to extend the water supply duration by a factor of two, compared to using the emergency generators only for the water network, while additionally almost half of the electricity demands can be resupplied. Algorithmic scaling is validated with a combination of the IEEE 123-bus test feeder and the D-Town water network.

      November 2022  25th International Conference on Intelligent Transportation Systems Conference

    Unsupervised Driving Event Discovery Based on Vehicle CAN-data

    Thomas Kreutz, Ousama Esbel, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1109/ITSC55140.2022.9922158

    Abstract
    The data collected from a vehicle’s Controller Area Network (CAN) can quickly exceed human analysis or annotation capabilities when considering fleets of vehicles, which stresses the importance of unsupervised machine learning methods. This work presents a simultaneous clustering and segmentation approach for vehicle CAN-data that identifies common driving events in an unsupervised manner. The approach builds on self-supervised learning (SSL) for multivariate time series to distinguish different driving events in the learned latent space. We evaluate our approach with a dataset of real Tesla Model 3 vehicle CAN-data and a two-hour driving session that we annotated with different driving events. With our approach, we evaluate the applicability of recent time series-related contrastive and generative SSL techniques to learn representations that distinguish driving events. Compared to state-of-the-art (SOTA) generative SSL methods for driving event discovery, we find that contrastive learning approaches reach similar performance.

      October 2022 Other

    Clustering of Threat Information to Mitigate Information Overload for Computer Emergency Response Teams

    Philipp Kuehn, Moritz Kerk, Marc Wendelborn, Christian Reuter

    PDF BibTeX

    Abstract
    The constantly increasing number of threats and the existing diversity of information sources pose challenges for Computer Emergency Response Teams (CERTs). In order to respond to new threats, CERTs need to gather information in a timely and comprehensive manner. However, the volume of information and sources can lead to information overload. This paper answers the question of how to reduce information overload for CERTs with the help of clustering methods. Conditions for such a framework were established and subsequently tested. In order to perform an evaluation, different types of evaluation metrics were introduced and selected in relation to the framework conditions. Furthermore, different vectorizations and distance measures in combination with the clustering methods were evaluated and interpreted. Two different ground-truth datasets were used for the evaluation, one containing threat messages and a dataset with messages from different news categories. The work shows that the K-means clustering method along with TF-IDF vectorization and cosine distance provide the best results in the domain of threat messages.

      October 2022  30th European Signal Processing Conference Conference

    False Discovery Rate Control for Grouped Variable Selection in High-Dimensional Linear Models Using the T-Knock Filter

    Jasin Machkour, Michael Muma, Daniel P. Palomar

    BibTeX DOI: 10.23919/EUSIPCO55093.2022.9909883

    Abstract
    High-dimensional variable selection is a challenging task, especially when groups of highly correlated variables are present in the data, such as in genomics research, direction-of-arrival estimation, and financial engineering. Recently, the T-Knock filter, a new framework for fast variable selection in high-dimensional settings has been developed. It provably controls the false discovery rate (FDR) at a given target level. However, its current version does not consider groups of highly correlated variables, which can lead to a loss in the true positive rate (TPR), i.e., the power. Hence, we propose the T-Knock+GVS filter that allows for grouped variable selection with FDR control in such settings. This is achieved by modifying the forward variable selection algorithm within the T- Knock filter and by adjusting the knockoff generation process such that the generated sets of knockoffs mimic the group correlation structure within the original set of variables. For a special case, we prove that the proposed T−Knock+GVS filter possesses the grouped variable selection property. Through a simulated high-dimensional genome-wide association study (GWAS), we show that the proposed method significantly increases the TPR, while controlling the FDR at the target level.

      October 2022  30th European Signal Processing Conference Conference

    Robust and Efficient Aggregation for Distributed Learning

    Stefan Vlaski, Christian Schroth, Michael Muma, Abdelhak M. Zoubir

    PDF BibTeX DOI: 10.23919/EUSIPCO55093.2022.9909822

    Abstract
    Distributed learning paradigms, such as federated and decentralized learning, allow for the coordination of models across a collection of agents, and without the need to exchange raw data. Instead, agents compute model updates locally based on their available data, and subsequently share the update model with a parameter server or their peers. This is followed by an aggregation step, which traditionally takes the form of a (weighted) average. Distributed learning schemes based on averaging are known to be susceptible to outliers. A single malicious agent is able to drive an averaging-based distributed learning algorithm to an arbitrarily poor model. This has motivated the development of robust aggregation schemes, which are based on variations of the median and trimmed mean. While such procedures ensure robustness to outliers and malicious behavior, they come at the cost of significantly reduced sample efficiency. This means that current robust aggregation schemes require significantly higher agent participation rates to achieve a given level of performance than their mean-based counterparts in non-contaminated settings. In this work we remedy this drawback by developing statistically efficient and robust aggregation schemes for distributed learning.

      October 2022  Die Architekt Article

    Zwischen Stadt und Land. Für eine neue Generation von Kulturlandschaften und Landschaftsräumen

    Julia Kemkemer-Böhmer, Annette Rudolph-Cleff

    PDF BibTeX

    Abstract
    Zwischen Stadt und Land werden nicht nur Abhängigkeiten, Verdrängungsprozesse und tiefe Einschnitte sichtbar, sondern auch Chancen, um auf die Herausforderungen des Klimawandels und der Urbanisierung zu antworten. Am Ende des Fortschrittsmythos ist es vielleicht möglich, über kulturelle Identitäten zwischen Stadt und Land nachzudenken, die nicht nur auf lokaler Tradition und individueller Erfahrung beruhen, sondern als zukunftsfähige Entwicklung im regionalen Wirtschaften und in kollektiver Verantwortung für den Natur- und Landschaftsraum gegründet sind. Welche Bilder haben wir für unseren Stadt- und Landschaftsraum?

      October 2022  12th IEEE Global Humanitarian Technology Conference (GHTC 2022) Conference

    Slums in Smart Cities - Rethink the Standard

    John Friesen, Martin Pietsch

    PDF BibTeX DOI: 10.1109/GHTC55712.2022.9911052

    Abstract
    Smart cities are often seen as a good way to meet the challenges of our time, such as population growth or climate change. The concepts to be developed in this context are often based on the assumption that the urban population is highly networked and adequately provided with infrastructure. At the same time, about one in four urban dwellers worldwide lives in a slum. Based on a literature review, we examine approaches that think smart cities and slums together. We show that smart cities approaches often do not take slums into account and that the standards assumed in these concepts should be rethought.

      October 2022  Radboud Universiteit Nijmegen Nijmegen Thesis

    A European Perspective on Crisis Informatics: Citizens’ and Authorities’ attitudes towards Social Media for Public Safety and Security

    Christian Reuter

    PDF BibTeX

    Abstract
    Mobilising helpers in the event of a flood or letting friends know that you are okay in the event of a terrorist attack – more and more people are using social media in emergency, crisis or disaster situations. Storms, floods, attacks or pandemics (esp. COVID-19) show that citizens use social media to inform themselves or to coordinate. This thesis presents qualitative and quantitative studies on the attitudes of emergency services and citizens in Europe towards social media in emergencies. Across the individual sub-studies, almost 10,000 people are surveyed including representative studies in the Netherlands, Germany, the UK and Italy. The work empirically shows that social media is increasingly important for emergency services, both for prevention and during crises; that private use of social media is a driving force in shaping opinions for organisational use; and that citizens have high expectations towards authorities, especially monitoring social media is expected, and sometimes responses within one hour. Depending on the risk culture, the data show further differences, e.g. whether the state (Germany) or the individual (Netherlands) is seen as primarily responsible for coping with the situation.

      October 2022  IEEE Robotics and Automation Letters Article

    A Modular and Portable Black Box Recorder for Increased Transparency of Autonomous Service Robots

    Max Schmidt, Jérôme Kirchhoff, Oskar von Stryk

    PDF BibTeX DOI: 10.1109/LRA.2022.3193633

    Abstract
    Autonomous service robots have great potential to support humans in tasks they cannot perform due to, amongst others, time constraints, work overload, or staff shortages. An important step for such service robots to be trusted or accepted by society is the provision of transparency. Its purpose is not only to communicate what a robot is doing according to the human interaction partners’ needs, it should also regard social and legal requirements. A black box recorder (inspired by flight recorders) increases the system’s transparency by facilitating the investigation of the cause of an incident, clarifying responsibilities, or improving the user’s understanding about the robot. In this work we propose the needed requirements of such a black box recorder for increased transparency of autonomous service robots, based on the related work. Further, we present a new modular and portable black box recorder design meeting these requirements. The applicability of the system is evaluated based on real-world robot data, using the realized open-source reference implementation.

      October 2022  Zeitschrift für Friedens- und Konfliktforschung Article

    Using Digitally Mediated Methods in Sensitive Contexts: A Threat Analysis and Critical Reflection on Security, Privacy, and Ethical Concerns in the Case of Afghanistan

    Laura Gianna Guntrum, Benjamin Güldenring, Franz Kuntke, Christian Reuter

    PDF BibTeX DOI: 10.1007/s42597-022-00088-2

    Abstract
    Given the lack of empirical examples of how research can be conducted via digital means in sensitive contexts, this paper provides a threat model using Afghanistan, where the Taliban took power in August 2021, as an example. Both technical and non-technical research-related risks are analyzed, paying attention to research ethics, data security, and privacy. We argue that any threat model and risk analysis is highly context-dependent. Our analysis reveals that in certain research processes, human security does not necessarily coincide with data security and that an ambivalence exists between privacy and usability. In addition to the concrete threat analysis, the paper identifies some general technical solutions (e.g., encryption methods, communication software) for different research steps to foster secure and ethically justifiable research.

      October 2022  Zeitschrift für Friedens- und Konfliktforschung Article

    Zur Schnittmenge von Informatik mit Friedens- und Sicherheitsforschung: Erfahrungen aus der interdisziplinären Lehre in der Friedensinformatik

    Christian Reuter, Thea Riebe, Jasmin Haunschild, Thomas Reinhold, Stefka Schmid

    BibTeX DOI: 10.1007/s42597-022-00078-4

    Abstract
    Interdisziplinäre Forschung und Lehre zwischen Informatik sowie Friedens- und Sicherheitsforschung ist vor dem Hintergrund, dass Konflikte im Cyberspace nicht mehr eine in der Zukunft liegende Fiktion, sondern eine realitätsnahe Möglichkeit darstellen, unabdingbar. Auch wenn zahlreiche etablierte Lehrveranstaltungen und Lehrbücher in der einen oder anderen Disziplin existieren, gilt dies nicht für deren Schnittmenge. Dieser Beitrag reflektiert die Einführung der in Bezug auf Thematik und Hörer*innenschaft interdisziplinären Lehrveranstaltung „Informationstechnologie für Frieden und Sicherheit“ für Studierende der Informatik, IT-Sicherheit und Wirtschaftsinformatik der Technischen Universität Darmstadt, sowie Friedens- und Konfliktforschung der TU Darmstadt in Kooperation mit der Goethe-Universität Frankfurt. Hierbei werden Herausforderungen und Lösungsansätze der interdisziplinären Lehre dargestellt und die Bedeutung dieser Lehre hervorgehoben.

      September 2022  24th International Conference on Human-Computer Interaction with Mobile Devices and Services Conference

    Comparing VR Exploration Support for Ground-Based Rescue Robots

    Julius von Willich, Andrii Matviienko, Sebastian Günther, Max Mühlhäuser

    BibTeX DOI: 10.1145/3528575.3551440

    Abstract
    Rescue robots have been extensively used in crisis situations for exploring dangerous areas. This exploration is usually facilitated via a remote operation by the rescue team. Although Virtual Reality (VR) was proposed to facilitate remote control due to its high level of immersion and situation awareness, we still lack intuitive and easy-to-use operation modes for search and rescue teams in VR environments. In this work, we propose four operation modes for ground-based rescue robots to utilize an efficient search and rescue: (a) Handle Mode, (b) Lab Mode, (c) Remote Mode, and (d) UI Mode. We evaluated these operation modes in a controlled lab experiment (N = 8) in terms of robot collisions, number of rescued victims, and mental load. Our results indicate that control modes with robot automation (UI and Remote mode) outperform modes with full control given to participants. In particular, we discovered that UI and Remote Mode lead to the lowest number of collisions, driving time, visible victims remaining, rescued victims, and mental load.

      September 2022  Sensors Article

    Activity-Free User Identification Using Wearables Based on Vision Techniques

    Alejandro Sanchez Guinea, Simon Heinrich, Max Mühlhäuser

    BibTeX DOI: 10.3390/s22197368

    Abstract
    In order to achieve the promise of smart spaces where the environment acts to fulfill the needs of users in an unobtrusive and personalized manner, it is necessary to provide means for a seamless and continuous identification of users to know who indeed is interacting with the system and to whom the smart services are to be provided. In this paper, we propose a new approach capable of performing activity-free identification of users based on hand and arm motion patterns obtained from an wrist-worn inertial measurement unit (IMU). Our approach is not constrained to particular types of movements, gestures, or activities, thus, allowing users to perform freely and unconstrained their daily routine while the user identification takes place. We evaluate our approach based on IMU data collected from 23 people performing their daily routines unconstrained. Our results indicate that our approach is able to perform activity-free user identification with an accuracy of 0.9485 for 23 users without requiring any direct input or specific action from users. Furthermore, our evaluation provides evidence regarding the robustness of our approach in various different configurations

      September 2022  Mensch und Computer 2022: Facing Realities Conference

    Perceptions and Use of Warning Apps - Did Recent Crises Lead to Changes in Germany?

    Christian Reuter, Jasmin Haunschild, Marc-André Kaufhold

    BibTeX DOI: 10.1145/3543758.3543770

    Abstract
    Warning and emergency apps are an integral part of crisis informatics and particularly relevant in countries that currently do not have cell broadcast, such as Germany. Previous studies have shown that such apps are regarded as relevant, but only around 16% of German citizens used them in 2017 and 2019. With the COVID-19 pandemic and a devastating flash flood, Germany has recently experienced severe crisis-related losses. By comparing data from representative surveys from 2017, 2019 and 2021, this study investigates whether these events have changed the perceptions of warning apps and their usage patterns in Germany. The study shows that while multi-hazard emergency and warning apps have been easily surpassed in usage by COVID-19 contact tracing apps, the use of warning apps has also increased and the pandemic has added new desired features. While these have been little-used during the COVID-19 pandemic, especially non-users see smartphone messengers app channels as possible alternatives to warning apps. In addition, regional warning apps appear promising, possibly because they make choosing a warning app easier when there are several available on the market.

      September 2022 Conference

    Mensch und Computer 2022: Facing Realities - Tagungsband

    PDF BibTeX DOI: 10.1145/3543758

    Abstract
    Die 2001 ins Leben gerufene Konferenz „Mensch und Computer“ (MuC) ist die größte Tagungsreihe der Mensch-Computer-Interaktion in Europa. Teilnehmenden aus Wissenschaft und Industrie bietet die MuC eine Plattform für Beiträge und Diskussionen zu innovativen Formen der Interaktion zwischen Menschen und digitaler Technik, zu nutzerorientierten Entwicklungsmethoden, interaktiven Anwendungen und weiteren Themen aus dem Spannungsfeld zwischen Nutzenden, Teams und Communities einerseits sowie den genutzten Informations- und Kommunikationstechnologien andererseits. Ziel der Tagung ist es, innovative Forschungsergebnisse zu diskutieren, den Informationsaustausch zwischen Wissenschaft und Praxis zu fördern, Forschungsaktivitäten und Ausbildung anzuregen sowie Wissenschaft, Praxis und Öffentlichkeit für die Relevanz nutzungs- und aufgabengerechter Technikgestaltung zu sensibilisieren. Hauptsächlich englischsprachige Fachbeiträge und deren Veröffentlichung in der „ACM Digital Library“ bzw. Digitalen Bibliothek der GI2 fördern die weltweite Sichtbarkeit der wissenschaftlichen Resultate der MuC.

      September 2022  Zeitschrift für Außen- und Sicherheitspolitik Article

    Book Review: Artificial Intelligence and The Future of Warfare: The USA China and Strategic Stability. Johnson,J. Manchester: University Press. (2021)

    Thomas Reinhold, Christian Reuter

    PDF BibTeX DOI: 10.1007/s12399-022-00918-4

    Abstract
    Technological and scientific progress, especially the rapid development of information technology (IT), plays a crucial role in peace and security issuesFootnote 1. Artificial Intelligence (AI) is one example. AI is a sub-discipline of computer science, dealing with computer systems capable of performing tasks which require human intelligenceFootnote 2. According to James Johnson, PhD, Lecturer in Strategic Studies in the Department of Politics & International Relations at the University of Aberdeen and author of the book Artificial Intelligence and the Future of Warfare, the hype around this has made it easy to overstate the opportunities and challenges posed by the development and deployment of AI in the military sphere. The author argues, that “speculations about super intelligent AI or the threat of superman AI to humanity” are entirely disconnected from today’s capabilities of AI. The book aims to address this problem by deciphering “proven capabilities and applications from mere speculation”, with a strong focus on the challenges AI poses to strategic stability, nuclear deterrence and how AI might influence nuclear weapon systems. The author concludes with implications and policy recommendations on how states could manage the escalatory risks posed by AI.

      September 2022  44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society Conference

    Fast and Sample Accurate R-Peak Detection for Noisy ECG Using Visibility Graphs

    Taulant Koka, Michael Muma

    BibTeX DOI: 10.1109/EMBC48229.2022.9871266

    Abstract
    More than a century has passed since Einthoven laid the foundation of modern electrocardiography and in recent years, driven by the advance of wearable and low budget devices, a sample accurate detection of R-peaks in noisy ECG-signals has become increasingly important. To accommodate these demands, we propose a new R-peak detection approach that builds upon the visibility graph transformation, which maps a discrete time series to a graph by expressing each sample as a node and assigning edges between intervisible samples. The proposed method takes advantage of the high connectivity of large, isolated values to weight the original signal so that R-peaks are amplified while other signal components and noise are suppressed. A simple thresholding procedure, such as the widely used one by Pan and Tompkins, is then sufficient to accurately detect the R-peaks. The weights are computed for overlapping segments of equal size and the time complexity is shown to be linear in the number of segments. Finally, the method is benchmarked against existing methods using the same thresholding on a noisy and sample accurate database. The results illustrate the potential of the proposed method, which outperforms common detectors by a significant margin.

      September 2022  ACM Transactions on Sensor Networks Article

    LoRaWAN Security: An Evolvable Survey on Vulnerabilities, Attacks and their Systematic Mitigation

    Frank Hessel, Lars Almon, Matthias Hollick

    BibTeX DOI: 10.1145/3561973

    Abstract
    The changing vulnerability and threat landscape constantly challenge the security of wireless communication standards and protocols. For the Internet of Things (IoT), LoRaWAN is one of the dominant technologies for urban environments, industrial settings, or critical infrastructures due to its low-power and long-range capabilities. LoRaWAN IoT deployments are expected to operate for multiple years or even decades. Hence, it is imperative to maintain operational security at all times while continuously evolving the security of the standard and its implementations. We survey LoRaWAN security and follow a systematic and evolvable approach that can be dynamically updated. To this end, we propose a novel methodology to create evolvable surveys which relate the analyzed critical security concepts, thus allowing IT security experts to reason about LoRaWAN security properties over time. With this, we provide a tool to hardware manufacturers, software developers and providers, and network operators to achieve sustainable security for IoT deployments.

      September 2022  The 39th International Conference on Machine Learning Conference

    IGLUE: A Benchmark for Transfer Learning across Modalities, Tasks, and Languages

    Emanuele Bugliarello, Fangyu Liu, Jonas Pfeiffer, Siva Reddy, Desmond Elliott, Edoardo M. Ponti, Ivan Vulić

    PDF BibTeX

    Abstract
    Reliable evaluation benchmarks designed for replicability and comprehensiveness have driven progress in machine learning. Due to the lack of a multilingual benchmark, however, vision-and-language research has mostly focused on English language tasks. To fill this gap, we introduce the Image-Grounded Language Understanding Evaluation benchmark. IGLUE brings together — by both aggregating pre-existing datasets and creating new ones — visual question answering, cross-modal retrieval, grounded reasoning, and grounded entailment tasks across 20 diverse languages. Our benchmark enables the evaluation of multilingual multimodal models for transfer learning, not only in a zero-shot setting, but also in newly defined few-shot learning setups. Based on the evaluation of the available state-of-the-art models, we find that translate-test transfer is superior to zero-shot transfer and that few-shot learning is hard to harness for many tasks. Moreover, downstream performance is partially explained by the amount of available unlabelled textual data for pretraining, and only weakly by the typological distance of target – source languages. We hope to encourage future research efforts in this area by releasing the benchmark to the community.

      September 2022  International Journal of Critical Infrastructure Protection Article

    The Digital Divide in State Vulnerability to Submarine Communications Cable Failure

    Jonas Franken, Thomas Reinhold, Lilian Reichert, Christian Reuter

    BibTeX DOI: 10.1016/j.ijcip.2022.100522

    Abstract
    The backbone network of submarine communication cables (SCC) carries 98% of international internet traffic. Coastal and island states strongly depend on this physical internet infrastructure to provide internet connectivity. Although about 100 SCC breakdowns of human or natural origin occur at yearly average, a literature review reveals that there is no approach to assess individual state vulnerability to SCC failure in global comparison. In this article, the global SCC network is modeled based on publicly available data. Besides the analysis of the global network properties, a focus is put on remaining bandwidth capacities in three different failure scenario simulations of SCC breakdowns. As a result, this study identifies 15 highly vulnerable states and overseas territories, and another 28 territories that are classified as partially vulnerable to SCC failures. Since economic market decisions shape the structure of the SCC network, an uneven distribution of redundancies and the resulting vulnerability of disadvantaged economies can be confirmed. Therefore, the study’s findings may contribute to a better assessment of the necessity of preventive protection measures of critical telecommunication infrastructures in states and territories characterized by high and medium vulnerability.

      September 2022  IEEE Transactions on Technology and Society Article

    Towards a Cyber Weapons Assessment Model – Assessment of the Technical Features of Malicious Software

    Thomas Reinhold, Christian Reuter

    BibTeX DOI: 10.1109/TTS.2021.3131817

    Abstract
    The revelation of the Stuxnet malware in 2010 shed light on the presence of state actors that are willing and capable of developing and using highly sophisticated, specialized malicious software for their political interests. These tools -often dubbed cyber weapons -are expected to become the next major advancement in weaponry technology. Besides the threats of offensive cyber operations for civil IT systems due to the interconnected nature of the cyberspace, international regulation of cyber weapons is -among other aspects -hindered by the fact that the military development and the strategic and tactical deployment of cyber weapons differ significantly from other weapons technologies. In order to establish measures of cyber arms related control treaties, it is crucial to identify these particular characteristics. Based on this premise, the article analyzes the current perspectives on cyber weapons, identifying their weaknesses of being either based on assumptions about adversarial actors or being applicable only after the usage of a malicious tool. In contrast to these approaches, the article focuses on the specific functional aspects of malware and presents an indicator-based assessment model based on parameters that can be measured prior to the application of malicious software. This enables the categorization of malicious tools as cyber weapons. Besides this, the article aims to introduce thought-provoking impulses with regard to social responsibility in computer science.

      August 2022  47th Conference on Local Computer Networks Conference

    ForestEdge: Unobtrusive Mechanism Interception in Environmental Monitoring

    Patrick Lampe, Markus Sommer, Artur Sterz, Jonas Höchst, Christian Uhl, Bernd Freisleben

    BibTeX DOI: 10.1109/LCN53696.2022.9843426

    Abstract
    A network for environmental monitoring typically requires a large number of sensors. If a longer service life is intended, it is essential that the deployed sensor systems can be upgraded without modifying hardware. Often, these networks rely on proprietary hardware/software components tailored to the desired functionality, but these could technically also be used for other applications. We present a demo of mechanism interception, a novel approach to unobtrusively add or modify the functionality of an existing networked system, in our case a TreeTalker, without touching any proprietary components. We demonstrate how a cloud infrastructure can be unobtrusively replaced by an edge infrastructure in a wireless sensor network. Our results indicate that mechanism interception is a compelling approach for our scenario to provide previously unavailable functionality without modifying existing components.

      August 2022  2022 IEEE International Conference on Communications Conference

    A Global Optimization Method for Energy-Minimal UAV-Aided Data Collection over Fixed Flight Path

    Guangping Lu, Jing Zhang, Lin Xiang, Xiaohu Ge

    PDF BibTeX DOI: 10.1109/ICC45855.2022.9838554

    Abstract
    This paper considers optimal resource allocation for data collection from multiple ground devices (GDs) using a rotary-wing unmanned aerial vehicle (UAV). The UAV’s flight path, i.e., the sequence of moving positions, is given a priori due to requirements of e.g. patrol and inspection missions, whereas the UAV’s trajectory, i.e., the path and time schedule of movement, remains dependent on its hovering positions and flying speeds along the path. To improve the spectral and energy efficiency of the GDs, the UAV employs a directional antenna and performs wireless power transfer (WPT) to the GDs before collecting data from them. We jointly optimize the UAV’s flying speeds, hovering locations, and radio resource allocation (including time, bandwidth and transmit power) for minimization of the total energy consumption of the UAV required for completing data collection along the flying path. We show that given any flight path, the propulsion energy consumption of the UAV is a convex function of the flight speeds. However, due to the highly directive transmission, communication and flight of the UAV become strongly coupled and complicates the problem, e.g. the selection of the UAV’s hovering points will affect both the order of serving the GDs and the antenna gain of the UAV. Moreover, nonconvexity in the flight path constraints further obscures an efficient solution to the resource allocation problem. To tackle these challenges, we propose an iterative algorithm based on the branch-and-bound (BnB) method, which can obtain the globally optimal solution when the flight path coincides with the boundary of a convex set. Simulation results show that compared with several baseline algorithms, the proposed algorithm can significantly lower the energy consumption of the UAV during data collection.

      August 2022  2022 IEEE International Conference on Communications Conference

    Delay- and Incentive-Aware Crowdsensing: A Stable Matching Approach for Coverage Maximization

    Bernd Simon, Sumedh Dongare, Tobias Mahn, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX DOI: 10.1109/ICC45855.2022.9838603

    Abstract
    Mobile crowdsensing (MCS) is a novel approach to increase the coverage, lower the costs, and increase the accuracy of sensing data. Its main idea is to collect sensor data using mobile units (MUs). The sensing is controlled by a mobile crowdsensing platform (MCSP) through the assignment of delay-sensitive sensing tasks to the MUs. Although promising, research effort in MCS is still needed to find task assignment solutions that maximize the coverage while considering the cost incurred by the MCSPs, the preferences of the MUs and the limited communication resources available. Specifically, we identify two main challenges: (i) A task assignment problem which incorporates the MCSP’s utility and the preferences of the MUs. (ii) An underlying communication resource allocation problem formulating the requirement of the timely transmission of sensing results given the limited communication resources. To address these challenges, we propose a novel two-stage matching algorithm. In the first stage, potential MU-task pairs are constructed considering the preferences of the MUs and the utility of the MCSP. In the second stage, the communication resource allocation is done based on potential MU-task pairs from the first stage. Through numerical simulations, we show that our proposed approach outperforms state-of-the-art methods in terms of the MCSP’s utility, coverage and MU’s satisfaction.

      August 2022  IEEE/ACM Transactions on Networking Article

    Multi-Stakeholder Service Placement via Iterative Bargaining With Incomplete Information

    Artur Sterz, Patrick Felka, Bernd Simon, Sabrina Klos, Anja Klein, Oliver Hinz, Bernd Freisleben

    PDF BibTeX DOI: 10.1109/TNET.2022.3157040

    Abstract
    Mobile edge computing based on cloudlets is an emerging paradigm to improve service quality by bringing computation and storage facilities closer to end users and reducing operating cost for infrastructure providers (IPs) and service providers (SPs). To maximize their individual benefits, IP and SP have to reach an agreement about placing and executing services on particular cloudlets. We show that a Nash Bargaining Solution (NBS) yields the optimal solution with respect to social cost and fairness if IP and SP have complete information about the parameters of their mutual cost functions. However, IP and SP might not be willing or able to share all information due to business secrets or technical limitations. Therefore, we present a novel iterative bargaining approach without complete mutual information to achieve substantial cost reductions for both IP and SP. Furthermore, we investigate how different degrees of information sharing impact social cost and fairness of the different approaches. Our evaluation based on the mobile augmented reality game Ingress shows that our approach achieves up to about 82% of the cost reduction that the NBS achieves and a cost reduction of up to 147% compared to traditional Take-it-or-Leave-it approaches, despite incomplete information.

      July 2022  SIGGRAPH ‘22: Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference

    Immersive-Labeler: Immersive Annotation of Large-Scale 3D Point Clouds in Virtual Reality

    Achref Doula, Tobias Güdelhöfer, Andrii Matviienko, Max Mühlhäuser, Alejandro Sanchez Guinea

    BibTeX DOI: 10.1145/3532719.3543249

    Abstract
    We present Immersive-Labeler, an environment for the annotation of large-scale 3D point cloud scenes of urban environments. Our concept is based on the full immersion of the user in a VR-based environment that represents the 3D point cloud scene while offering adapted visual aids and intuitive interaction and navigation modalities. Through a user-centric design, we aim to improve the annotation experience and thus reduce its costs. For the preliminary evaluation of our environment, we conduct a user study (N=20) to quantify the effect of higher levels of immersion in combination with the visual aids we implemented on the annotation process. Our findings reveal that higher levels of immersion combined with object-based visual aids lead to a faster and more engaging annotation process.

      July 2022  Global Media Journal - German Edition Article

    Remotely Accessing the Field and Building Trust with Distant Sources. Perspectives from Journalism Practice for Ethnographic Research

    Laura Gianna Guntrum, Mira Keßler, Jignesh Patel, Anna Varfolomeeva

    PDF BibTeX DOI: 10.22032/dbt.53026

    Abstract
    Journalists and ethnographic researchers, such as anthropologists, sociologists or media scholars, have comparable ways of establishing initial contacts with people from their fields of interest. Due to the ongoing COVID-19 pandemic and consequential travel restrictions and social distancing, it has become increasingly difficult to access a field. Taking inspiration from social anthropologist Ulf Hannerz (2004, p. 226), who compared journalists and anthropologists as “neighboring groups engaged in a somehow parallel pursuit,” this article explores what researchers may learn from practitioners who conduct research without being on-site. Fed by various practical journalists’ experiences, the article aims to investigate how information and communication technologies (ICTs) and digitally mediated methods, such as online search tools and social media, can be used to establish contacts and gain trust remotely. Here, the relevance of these methods for accessing a field in general goes beyond the limitations imposed during the COVID-19 pandemic and can be of interest to all those who face difficulties of field access of any kind. Ultimately, this article reflects on corresponding ethical challenges that may arise while conducting research remotely.

      July 2022  Australasian Conference on Information Systems 2022 Conference

    A Design Science Artefact for Cyber Threat Detection and Actor Specific Communication

    Kaan Eyilmez, Ali Sercan Basyurt, Stefan Stieglitz, Christoph Fuchss, Marc-André Kaufhold, Christian Reuter, Milad Mirabaie

    PDF BibTeX

    Abstract
    Over the past few decades, the number and variety of cyberattacks and malware patterns have increased immensely. As a countermeasure, computer emergency response teams were established with the responsibility of securing the cyber environment. However, recent studies revealed that currently performed manual processes and the unavailability of adequate tools impede the achievement of cybersecurity. To address these challenges, we followed the Design Science paradigm to develop an artefact that improves the evaluation of open-source intelligence obtained from Twitter as well as the actor-specific communication of cyber threat information. Subsequently, the implemented artefact will be evaluated through semi-structured interviews with subject matter experts. This research in progress article presents the identified research gap and describes the development process and the endeavor to contribute to the cybersecurity domain theoretically with design principles for the development of an instrument and practically by implementing an artefact that supports domain experts in their work.

      July 2022  Academy of Management proceedings Article

    Sustainable aim or personal gain? Effects of Personal and Sustainable Value on Crowdfunding Success

    Sven Siebeneicher, Carolin Bock

    PDF BibTeX DOI: 10.5465/AMBPP.2022.14306abstract

    Abstract
    We extend the entrepreneurship literature on crowdfunding by investigating how the relation between personal value and shared sustainable value affects crowdfunding success. To define shared sustainable value, we disaggregate sustainability into three interrelated dimensions: ecologic value, economic value, and social value. Relying on signaling theory, we identify the value proposed in campaign teasers and descriptions by deriving and utilizing reliable word lists for text analysis. Our findings suggest that “first impression matters”. Teasers and descriptions both have significant effects, and crowdfunding can be employed to foster altruistic behavior. For our analysis, we rely on a sample of 45,608 Kickstarter campaigns.

      July 2022  Academy of Management Proceedings Article

    Flip the Tweet - The Two-Sided Coin of Entrepreneurial Empathy

    Carolin Bock, Konstantin Kurz

    PDF BibTeX DOI: 10.5465/AMBPP.2022.14076abstract

    Abstract
    Is empathy merely a good thing for entrepreneurs? In contrast to the hitherto predominantly positive view in entrepreneurship literature, psychology and management scholars have recently adopted a more critical perspective. Transferring their findings, we provide a novel “too-much-of-a-good-thing” perspective on entrepreneurial empathy. We test our model against a dataset of 4,725 real entrepreneurs and demonstrate that empathy influences opportunity recognition, evaluation and exploitation in an inverted U-shaped pattern. We also show that these exploited opportunities then lead to higher amounts of achieved financial resources. These findings provide strong evidence for considering entrepreneurial empathy an important but highly ambiguous success factor.

      July 2022  44th International Conference on Software Engineering Conference

    Change is the Only Constant: Dynamic Updates for Workflows

    Daniel Sokolowski, Pascal Weisenburger, Guido Salvaneschi

    BibTeX DOI: 10.1145/3510003.3510065

    Abstract
    Software systems must be updated regularly to address changing requirements and urgent issues like security-related bugs. Traditionally, updates are performed by shutting down the system to replace certain components. In modern software organizations, updates are increasingly frequent—up to multiple times per day—hence, shutting down the entire system is unacceptable. Safe dynamic software updating (DSU) enables component updates while the system is running by determining when the update can occur without causing errors. Safe DSU is crucial, especially for long-running or frequently executed asynchronous transactions (workflows), e.g., user-interactive sessions or order fulfillment processes. Unfortunately, previous research is limited to synchronous transaction models and does not address this case.In this work, we propose a unified model for safe DSU in workflows. We discuss how state-of-the-art DSU solutions fit into this model and show that they incur significant overhead. To improve the performance, we introduce Essential Safety, a novel safe DSU approach that leverages the notion of non-essential changes, i.e., semantics preserving updates. In 106 realistic BPMN workflows, Essential Safety reduces the delay of workflow completions, on average, by 47.8 compared to the state of the art. We show that the distinction of essential and non-essential changes plays a crucial role in this reduction and that, as suggested in the literature, non-essential changes are frequent: at least 60 and often more than 90 of systems’ updates in eight monorepos we analyze.

      July 2022  European Journal for Security Research Article

    U.S. Security Policy: The Dual-Use Regulation of Cryptography and its Effects on Surveillance

    Thea Riebe, Philipp Kuehn, Philipp Imperatori, Christian Reuter

    PDF BibTeX DOI: 10.1007/s41125-022-00080-0

    Abstract
    Cryptography has become ubiquitous in communication technology and is considered a necessary part of information security. However, both the regulation to restrict access to cryptography, as well as practices to weaken or break encryption, are part of the States’ security policies. The United States (U.S.) regulate cryptography for export in international trade as a dual-use good. However, the regulation has been increasingly loosened and transferred to bilateral agreements with Information and Communication Technology companies. At the same time, the National Security Agency attempted to implement a government encryption standard to guarantee itself easier access to data, thus progressively expanding surveillance on non-U.S. citizens. In this paper, using comparative policy analysis, we examine the evolution of both security policies by tracing the historical development of U.S. regulation of cryptography as a dual-use good, and surveillance technologies, and practices used from the 1990s to today. We conclude that the impact of the dual-use regulation has affected the efficiency of surveillance technology, by loosening regulations only for mass communication services, thereby supporting the proliferation of surveillance intermediaries, while working on strategies to collaborate and exploit their coverage.

      July 2022  IEEE Transactions on Vehicular Technology Article

    Scheduling for Massive MIMO with Hybrid Precoding using Contextual Multi-Armed Bandits

    Weskley V. F. Mauricio, Tarcisio Ferreira Maciel, Anja Klein, Francisco Rafael Marques Lima

    PDF BibTeX DOI: 10.1109/TVT.2022.3166654

    Abstract
    In this work we study different scheduling problems in the downlink of a Frequency Division Duplex multiuser wireless system that employs a hybrid precoding antenna architecture for massive Multiple Input Multiple Output. In this context, we propose a scheduling framework using Reinforcement Learning (RL) tools, namely Contextual Multi-Armed Bandits (CMAB), that can dynamically adapt themselves to solve three scheduling problems, which are: i) Maximum Throughput (MT); ii) Maximum Throughput with Fairness Guarantees (MTFG), and; iii) Maximum Throughput with QoS Guarantees (MTQG), which are well-known relevant problems. Before performing scheduling itself, we exploit statistical Channel State Information (CSI) to create clusters of spatially compatible User Equipmentss (UEss). This structure, combined with the usage of Zero-Forcing precoding, allows us to reduce the scheduler complexity by considering each cluster as an independent virtual RL scheduling agent. Next, we apply a new learning-based scheduler aiming to optimize the desired system performance metric. Moreover, only scheduled UEss need to feed back instantaneous equivalent CSI, which also reduces the signaling overhead of the proposal. The superiority of the proposed framework is demonstrated through numerical simulations in comparison with reference solutions.

      June 2022  36th European Conference on Object-Oriented Programming Conference

    Functional Programming for Distributed Systems with XC

    Giorgio Audrito, Roberto Casadei, Ferruccio Damiani, Guido Salvaneschi, Mirko Viroli

    PDF BibTeX DOI: 10.4230/LIPIcs.ECOOP.2022.20

    Abstract
    Programming distributed systems is notoriously hard due to - among the others - concurrency, asynchronous execution, message loss, and device failures. Homogeneous distributed systems consist of similar devices that communicate to neighbours and execute the same program: they include wireless sensor networks, network hardware, and robot swarms. For the homogeneous case, we investigate an experimental language design that aims to push the abstraction boundaries farther, compared to existing approaches. In this paper, we introduce the design of XC, a programming language to develop homogeneous distributed systems. In XC, developers define the single program that every device executes and the overall behaviour is achieved collectively, in an emergent way. The programming framework abstracts over concurrency, asynchronous execution, message loss, and device failures. We propose a minimalistic design, which features a single declarative primitive for communication, state management, and connection management. A mechanism called alignment enables developers to abstract over asynchronous execution while still retaining composability. We define syntax and operational semantics of a core calculus, and briefly discuss its main properties. XC comes with two DSL implementations: a DSL in Scala and one in C++. An evaluation based on smart-city monitoring demonstrates XC in a realistic application.

      June 2022 Other

    RIS assisted device activity detection with statistical channel state information

    Friedemann Laue, Vahid Jamali, Robert Schober

    PDF BibTeX DOI: 10.48550/arXiv.2206.06805

    Abstract
    This paper studies reconfigurable intelligent surface (RIS) assisted device activity detection for grant-free (GF) uplink transmission in wireless communication networks. In particular, we consider mobile devices located in an area where the direct link to an access point (AP) is blocked. Thus, the devices try to connect to the AP via a reflected link provided by an RIS. Therefore, a RIS phase-shift design is desired that covers the entire blocked area with a wide reflection beam because the exact locations and times of activity of the devices are unknown in GF transmission. In order to study the impact of the phase-shift design on the device activity detection, we derive a generalized likelihood ratio test (GLRT) based detector and present an analytical expression for the probability of detection. Assuming knowledge of statistical CSI, we formulate an optimization problem for the phase-shift design for maximization of the guaranteed probability of detection for all locations within a given coverage area. To tackle the non-convexity of the problem, we propose two different approximations of the objective function. The first approximation leads to a design that aims to reduce the variations of the end-to-end channel while taking system parameters such as transmit power, noise power, and probability of false alarm into account. The second approximation can be adopted for versatile RIS deployments because it only depends on the line-of-sight component of the end-to-end channel and is not affected by system parameters. For comparison, we also consider a phase-shift design maximizing the average channel gain and a baseline analytical phase-shift design for large blocked areas. Our performance evaluation shows that the proposed approximations result in phase-shift designs that guarantee high probability of detection across the coverage area and outperform the baseline designs.

      June 2022  Fluid Power: Digital, Reliable, Sustainable - 13th International Fluid Power Conference Conference

    Data Management as an Enabler of Sustainability – Discussion Using the Example of a Digital Data Sheet

    Kevin Logan, Marvin Meck, N. Preuß, Philipp Wetterich, Peter F. Pelz

    PDF BibTeX DOI: 10.18154/RWTH-2023-04624

    Abstract
    Sustainable systems require sustainable data. Imposing the paradigm of sustainability on data corresponds to making it findable, accessible, interoperable and reusable (FAIR). This is vital for planning a fluid system fulfilling functionality as well as satisfactory quality. Therefore, customers are dependent on component suppliers making FAIR product data available. Rather than assuming the context of runtime machine to machine communication and service-oriented interfaces, i.e., solutions based on digital twins, the presented work proposes a digital data sheet for components based on citation of information models compliant with W3C standards. This enables suppliers to provide comprehensible and transparent data and customers to implement sustainability already during the design process of fluid systems.

      June 2022  Datenschutz und Datensicherheit Article

    Kollaboration im Datenschutz: Collaboration Engineering als Instrument zur partizipativen und nachhaltigen Gestaltung von Datenschutzprozessen

    Gerrit Hornung, Matthias Söllner, Jan-Philipp Stroscher, Eva-Maria Zahn

    PDF BibTeX DOI: 10.1007/s11623-022-1625-4

    Abstract
    Die DSGVO enthält sehr unterschiedliche Vorgaben zur Zusammenarbeit verschiedener Akteure, aber kein systematisches Modell für die Ausgestaltung dieser Formen der Zusammenarbeit. Gerade für KMU könnten standardisierte Verfahren ein Ansatz zur effizienten Umsetzung sein. Der Beitrag untersucht die Frage, an welchen Stellen Ansätze des Collaboration Engineering Zusammenarbeitsprozesse ermöglichen würden, die die gesetzlichen Vorgaben mit Leben füllen.

      May 2022  60th Annual Meeting of the Association for Computational Linguistics Conference

    UKP-SQUARE: An Online Platform for Question Answering Research

    Tim Baumgärtner, Kexin Wang, Rachneet Sachdeva, Gregor Geigle, Max Eichler, Clifton Poth, Hannah Sterz, Haritz Puerto, Leonardo F. R. Ribeiro, Jonas Pfeiffer, Nils Reimers, Gözde Gül Şahin, Iryna Gurevych

    PDF BibTeX

      May 2022  10th Edition of the International Conference on Networked Systems (NETYS 2022) Conference

    Bird@Edge: Bird Species Recognition at the Edge

    Jonas Höchst, Hicham Bellafkir, Patrick Lampe, Markus Vogelbacher, Markus Mühling, Daniel Schneider, Kim Lindner, Sascha Rösner, Dana G. Schabo, Nina Farwig, Bernd Freisleben

    PDF BibTeX DOI: 10.1007/978-3-031-17436-0_6

    Abstract
    We present Bird@Edge, an Edge AI system for recognizing bird species in audio recordings to support real-time biodiversity monitoring. Bird@Edge is based on embedded edge devices operating in a distributed system to enable efficient, continuous evaluation of soundscapes recorded in forests. Multiple ESP32-based microphones (called Bird@Edge Mics) stream audio to a local Bird@Edge Station, on which bird species recognition is performed. The results of several Bird@Edge Stations are transmitted to a backend cloud for further analysis, e.g., by biodiversity researchers. To recognize bird species in soundscapes, a deep neural network based on the EfficientNet-B3 architecture is trained and optimized for execution on embedded edge devices and deployed on a NVIDIA Jetson Nano board using the DeepStream SDK. Our experiments show that our deep neural network outperforms the state-of-the-art BirdNET neural network on several data sets and achieves a recognition quality of up to 95.2% mean average precision on soundscape recordings in the Marburg Open Forest, a research and teaching forest of the University of Marburg, Germany. Measurements of the power consumption of the Bird@Edge components highlight the real-world applicability of the approach. All software and firmware components of Bird@Edge are available under open source licenses.

      May 2022  22nd Power Systems Computation Conference (PSCC 2022) Conference

    The Water Energy Nexus: Improved Emergency Grid Restoration with DERs

    Martin Pietsch, Florian Steinke

    PDF BibTeX

    Abstract
    Water networks as critical infrastructures typically feature emergency electricity generators for bridging short power blackouts. We propose to combine these black start capable generators with available distributed energy resources (DERs) in the power grid, often photovoltaic generation, to jointly restore both the electricity and the water grid after blackouts. This is mutually beneficial for both networks since common grid- following inverters of DERs cannot supply power without a grid- forming nucleus. We model both grids as a coupled graph and formulate a stochastic mixed-integer linear program to determine optimal switch placement and/or switching sequences. Limited fuel and power availabilities, grid-forming constraints, storages, and an even distribution of available resources are considered. By minimizing the number of switching devices and switching events we target manual operability. The proposed method extends the time that can be bridged until a full restoration of the main power grid is achieved. For a small example, we demonstrate that given enough solar radiation our solution can double the water supply duration compared to using the generators only for the water network, while additionally resupplying almost half of the electricity demand. Algorithmic scaling is validated with a combination of the IEEE 123-bus test feeder and the D-Town water network.

      May 2022  Transactions of the Association for Computational Linguistics Article

    Retrieve Fast, Rerank Smart: Cooperative and Joint Approaches for Improved Cross-Modal Retrieval

    Gregor Geigle, Jonas Pfeiffer, Nils Reimers, Ivan Vulić, Iryna Gurevych

    PDF BibTeX DOI: 10.1162/tacl_a_00473

    Abstract
    Current state-of-the-art approaches to cross- modal retrieval process text and visual input jointly, relying on Transformer-based architectures with cross-attention mechanisms that attend over all words and objects in an image. While offering unmatched retrieval performance, such models: 1) are typically pretrained from scratch and thus less scalable, 2) suffer from huge retrieval latency and inefficiency issues, which makes them impractical in realistic applications. To address these crucial gaps towards both improved and efficient cross- modal retrieval, we propose a novel fine-tuning framework that turns any pretrained text-image multi-modal model into an efficient retrieval model. The framework is based on a cooperative retrieve-and-rerank approach that combines: 1) twin networks (i.e., a bi-encoder) to separately encode all items of a corpus, enabling efficient initial retrieval, and 2) a cross-encoder component for a more nuanced (i.e., smarter) ranking of the retrieved small set of items. We also propose to jointly fine- tune the two components with shared weights, yielding a more parameter-efficient model. Our experiments on a series of standard cross-modal retrieval benchmarks in monolingual, multilingual, and zero-shot setups, demonstrate improved accuracy and huge efficiency benefits over the state-of-the-art cross- encoders.

      May 2022  Transactions on Emerging Telecommunications Technologies Article

    LoRaWAN Security Issues and Mitigation Options by the Example of Agricultural IoT Scenarios

    Franz Kuntke, Vladimir Romanenko, Sebastian Linsner, Enno Steinbrink, Christian Reuter

    BibTeX DOI: 10.1002/ett.4452

    Abstract
    The Internet of Things (IoT) is a major trend that is seen as a great opportunity to improve efficiency in many domains, including agriculture. This technology could transform the sector, improving the management and quality of agricultural operations, for example, crop farming. The most promising data transmission standard for this domain seems to be Long Range Wide Area Network (LoRaWAN), a popular representative of low power wide area network technologies today. LoRaWAN, like any wireless protocol, has properties that can be exploited by attackers, which has been a topic of multiple research papers in recent years. By conducting a systematic literature review, we build a recent list of attacks, as well as collect mitigation options. Taking a look at a concrete use case (IoT in agriculture) allows us to evaluate the practicality of both exploiting the vulnerabilities and implementing the countermeasures. We detected 16 attacks that we grouped into six attack types. Along with the attacks, we collect countermeasures for attack mitigation. Developers can use our findings to minimize the risks when developing applications based on LoRaWAN. These mostly theoretical security recommendations should encourage future works to evaluate the mitigations in practice.

      April 2022  2022 CHI Conference on Human Factors in Computing Systems Conference

    BikeAR: Understanding Cyclists’ Crossing Decision-Making at Uncontrolled Intersections using Augmented Reality

    Andrii Matviienko, Florian Müller, Dominik Schön, Paul Seesemann, Sebastian Günther, Max Mühlhäuser

    BibTeX DOI: 10.1145/3491102.3517560

    Abstract
    Cycling has become increasingly popular as a means of transportation. However, cyclists remain a highly vulnerable group of road users. According to accident reports, one of the most dangerous situations for cyclists are uncontrolled intersections, where cars approach from both directions. To address this issue and assist cyclists in crossing decision-making at uncontrolled intersections, we designed two visualizations that: (1) highlight occluded cars through an X-ray vision and (2) depict the remaining time the intersection is safe to cross via a Countdown. To investigate the efficiency of these visualizations, we proposed an Augmented Reality simulation as a novel evaluation method, in which the above visualizations are represented as AR, and conducted a controlled experiment with 24 participants indoors. We found that the X-ray ensures a fast selection of shorter gaps between cars, while the Countdown facilitates a feeling of safety and provides a better intersection overview.

      April 2022  CHI ‘22: CHI Conference on Human Factors in Computing Systems Conference

    SkyPort: Investigating 3D Teleportation Methods in Virtual Environments

    Andrii Matviienko, Florian Müller, Martin Schmitz, Marco Fendrich, Max Mühlhäuser

    BibTeX DOI: 10.1145/3491102.3501983

    Abstract
    Teleportation has become the de facto standard of locomotion in Virtual Reality (VR) environments. However, teleportation with parabolic and linear target aiming methods is restricted to horizontal 2D planes and it is unknown how they transfer to the 3D space. In this paper, we propose six 3D teleportation methods in virtual environments based on the combination of two existing aiming methods (linear and parabolic) and three types of transitioning to a target (instant, interpolated and continuous). To investigate the performance of the proposed teleportation methods, we conducted a controlled lab experiment (N = 24) with a mid-air coin collection task to assess accuracy, efciency and VR sickness. We discovered that the linear aiming method leads to faster and more accurate target selection. Moreover, a combination of linear aiming and instant transitioning leads to the highest efciency and accuracy without increasing VR sickness.

      April 2022  2022 CHI Conference on Human Factors in Computing Systems Conference

    VR-Surv: A VR-Based Privacy Preserving Surveillance System

    Achref Doula, Alejandro Sanchez Guinea, Max Mühlhäuser

    BibTeX DOI: 10.1145/3491101.3519645

    Abstract
    The recent advances in smart city infrastructure have provided support for a higher adoption of surveillance cameras as a mainstream crime prevention measure. However, a consequent massive deployment raises concerns about privacy issues among citizens. In this paper, we present VR-Surv, a VR-based privacy aware surveillance system for large scale urban environments. Our concept is based on conveying the semantics of the scene uniquely, without revealing the identity of the individuals or the contextual details that might violate the privacy of the entities present in the surveillance area. For this, we create a virtual replica of the areas of interest, in real-time, through the combination of procedurally generated environments and markerless motion capture models. The results of our preliminary evaluation revealed that our system successfully conceals privacy-sensitive data, while preserving the semantics of the scene. Furthermore, participants in our user study expressed higher acceptance to being surveilled through the proposed system.

      April 2022  2022 IEEE Conference on Virtual Reality and 3D User Interfaces Conference

    Effects of the Level of Detail on the Recognition of City Landmarks in Virtual Environments

    Achref Doula, Philipp Kaufmann, Alejandro Sanchez Guinea, Max Mühlhäuser

    PDF BibTeX DOI: 10.1109/VRW55335.2022.00281

    Abstract
    The reconstruction of city landmarks is central to creating recognizable virtual environments representing real cities. Despite the recent advances, it is still not clear what level of detail (LOD) to adopt when reconstructing those landmarks for their correct recognition, and if particular architectural styles represent specific challenges in this respect. In this paper, we investigate the effect of LOD on landmark recognition, generally, and on some architectural styles, specifically. The results of our user study show that higher LOD lead to a better landmark identification. Particularly, Neoclassical-style buildings need more details to be individually distinguished from similar ones.

      April 2022  International Journal of Disaster Risk Science Article

    Resilience in Agriculture: Communication and Energy Infrastructure Dependencies of German Farmers

    Franz Kuntke, Sebastian Linsner, Enno Steinbrink, Jonas Franken, Christian Reuter

    PDF BibTeX DOI: 10.1007/s13753-022-00404-7

    Abstract
    Agriculture is subject to high demands regarding resilience as it is an essential component of the food production chain. In the agricultural sector, there is an increasing usage of digital tools that rely on communication and energy infrastructures. Should disruption occur, such strengthened dependencies on other infrastructures increase the probability of ripple effects. Thus, there is a need to analyze the resilience of the agricultural sector with a specific focus on the effects of digitalization. This study works out resilience capacities of the interconnected technologies used in farm systems based on the experiences and opinions of farmers. Information was gathered through focus group interviews with farmers (N = 52) and a survey with participants from the agricultural sector (N = 118). In particular, the focus is put on the digital tools and other information and communication technologies they use. Based on a definition of resilience capacities, we evaluate resilience regarding energy and communication demands in various types of farm systems. Especially important are the resilience aspects of modern systems’ digital communication as well as the poorly developed and nonresilient network infrastructure in rural areas that contrast with the claim for a resilient agriculture. The result is a low robustness capacity, as our analysis concludes with the risk of food production losses.

      April 2022  Physical Review E Article

    Motif-based mean-field approximation of interacting particles on clustered networks

    Kai Cui, Wasiur R. KhudaBukhsh, Heinz Koeppl

    PDF BibTeX DOI: 10.1103/PhysRevE.105.L042301

    Abstract
    Interacting particles on graphs are routinely used to study magnetic behavior in physics, disease spread in epidemiology, and opinion dynamics in social sciences. The literature on mean-field approximations of such systems for large graphs typically remains limited to specific dynamics, or assumes cluster-free graphs for which standard approximations based on degrees and pairs are often reasonably accurate. Here, we propose a motif-based mean-field approximation that considers higher-order subgraph structures in large clustered graphs. Numerically, our equations agree with stochastic simulations where existing methods fail.

      April 2022  10th International Conference on Learning Representations (ICLR 2022) Conference

    Learning Graphon Mean Field Games and Approximate Nash Equilibria

    Kai Cui, H. Koeppl

    BibTeX

    Abstract
    Recent advances at the intersection of dense large graph limits and mean field games have begun to enable the scalable analysis of a broad class of dynamical sequential games with large numbers of agents. So far, results have been largely limited to graphon mean field systems with continuous-time diffusive or jump dynamics, typically without control and with little focus on computational methods. We propose a novel discrete-time formulation for graphon mean field games as the limit of non-linear dense graph Markov games with weak interaction. On the theoretical side, we give extensive and rigorous existence and approximation properties of the graphon mean field solution in sufficiently large systems. On the practical side, we provide general learning schemes for graphon mean field equilibria by either introducing agent equivalence classes or reformulating the graphon mean field system as a classical mean field system. By repeatedly finding a regularized optimal control solution and its generated mean field, we successfully obtain plausible approximate Nash equilibria in otherwise infeasible large dense graph games with many agents. Empirically, we are able to demonstrate on a number of examples that the finite-agent behavior comes increasingly close to the mean field behavior for our computed equilibria as the graph or system size grows, verifying our theory. More generally, we successfully apply policy gradient reinforcement learning in conjunction with sequential Monte Carlo methods.

      March 2022  Science and Engineering Ethics Article

    Dual-Use and Trustworthy? A Mixed Methods Analysis of AI Diffusion between Civilian and Defense R&D

    Stefka Schmid, Thea Riebe, Christian Reuter

    BibTeX DOI: 10.1007/s11948-022-00364-7

    Abstract
    Artificial Intelligence (AI) seems to be impacting all industry sectors, while becoming a motor for innovation. The diffusion of AI from the civilian sector to the defense sector, and AI’s dual-use potential has drawn attention from security and ethics scholars. With the publication of the ethical guideline Trustworthy AI by the European Union (EU), normative questions on the application of AI have been further evaluated. In order to draw conclusions on Trustworthy AI as a point of reference for responsible research and development (R&D), we approach the diffusion of AI across both civilian and military spheres in the EU. We capture the extent of technological diffusion and derive European and German patent citation networks. Both networks indicate a low degree of diffusion of AI between civilian and defense sectors. A qualitative investigation of project descriptions of a research institute’s work in both civilian and military fields shows that military AI applications stress accuracy or robustness, while civilian AI reflects a focus on human-centric values. Our work represents a first approach by linking processes of technology diffusion with normative evaluations of R&D.

      March 2022  5th Workshop on System Software for Trusted Execution (SysTEX’22) Conference

    Always-trusted IoT - Making IoT Devices Trusted with Minimal Overhead

    Zsolt István, Paul Rosero, Philippe Bonnet

    PDF BibTeX

    Abstract
    Internet-of-Things (Iot) devices are becoming increasingly prevalent, with many of them not only relaying data to the Cloud but also being capable of local computation. This capability could be used for many purposes: detecting sensor tampering, compression or anonymization of data before uploading to the cloud, or even participating in distributed Machine Learning. IoT devices are not only at risk of malicious and misbehaving software, but due to their deployment in unprotected locations, they are also at risk of physical attackers and tampering. Even though there are many exciting local computation ideas, the authenticity of computations performed on most IoT devices cannot be guaranteed. In clouds, Trusted Execution Environments (TEEs) already offer trust in the computation carried out even in the presence of a physical attacker, without slowing applications down. In IoT devices, however, such TEEs introduce large performance overheads and increase energy consumption. In this project we propose a radical way forward: to design IoT platforms with processors that do not rely on off-chip memory and instead keep application state on on-chip memory that is easier to protect. This design reduces the overhead of TEEs significantly: it eliminates the cost of securing off-chip memory from attackers. It is important to note that, in addition to fresh thinking on how to design processors with more on-chip memory, computation will also have to be re-imagined to fit in a reduced memory footprint.

      February 2022 Other

    Nearest-Neighbor-based Collision Avoidance for Quadrotors via Reinforcement Learning

    R. Ourari, K. Cui, Ahmed A. Elshamanhory, H. Koeppl

    PDF BibTeX DOI: 10.48550/arXiv.2104.14912

    Abstract
    Collision avoidance algorithms are of central interest to many drone applications. In particular, decentralized approaches may be the key to enabling robust drone swarm solutions in cases where centralized communication becomes computationally prohibitive. In this work, we draw biological inspiration from flocks of starlings (Sturnus vulgaris) and apply the insight to end-to-end learned decentralized collision avoidance. More specifically, we propose a new, scalable observation model following a biomimetic nearest-neighbor information constraint that leads to fast learning and good collision avoidance behavior. By proposing a general reinforcement learning approach, we obtain an end-to-end learning-based approach to integrating collision avoidance with arbitrary tasks such as package collection and formation change. To validate the generality of this approach, we successfully apply our methodology through motion models of medium complexity, modeling momentum and nonetheless allowing direct application to real world quadrotors in conjunction with a standard PID controller. In contrast to prior works, we find that in our sufficiently rich motion model, nearest-neighbor information is indeed enough to learn effective collision avoidance behavior. Our learned policies are tested in simulation and subsequently transferred to real-world drones to validate their real-world applicability.

      February 2022  Cyber Security Politics: Socio-Technological Transformations and Political Fragmentation Other

    Cultural Violence and Fragmentation on Social Media: Interventions and Countermeasures by Humans and Social Bots

    Jasmin Haunschild, Marc-André Kaufhold, Christian Reuter

    BibTeX DOI: 10.4324/9781003110224-5

    Abstract
    As a prime example of socio-technological transformation, social media services exert an enormous impact on modern culture. They are nowadays widely established for everyday life uses, but also during natural and man-made crises and political conflicts. For instance, Facebook was part of the Arabic Spring, facilitating the communication and interaction between participants of political protests. However, social media is not only used for good: Based on the notions of cultural violence and cultural peace, this chapter shows the potential for political fragmentation through social media, focusing on fake news and terrorism propaganda and their amplified dissemination through social bots. The chapter shows that regarding both the problematic aspects and the countermeasures, technology plays the role of an amplifier, enabling effects such as astroturfing and smoke screening, but also enhancing social bot detection. This chapter examines what these socio-technological transformations through social media imply in terms of legitimacy and trust.

      February 2022  Computers & Security Article

    A Feature-driven Method for Automating the Assessment of OSINT Cyber Threat Sources

    Andrea Tundis, Samuel Ruppert, Max Mühlhäuser

    PDF BibTeX DOI: 10.1016/j.cose.2021.102576

    Abstract
    Global malware campaigns and large-scale data breaches show how everyday life can be impacted when the defensive measures fail to protect computer systems from cyber threats. Understanding the threat landscape and the adversaries’ attack tactics to perform it represent key factors for enabling an efficient defense against threats over the time. Of particular importance is the acquisition of timely and accurate information from threats intelligence sources available on the web which can provide additional intelligence on emerging threats even before they can be observed as actual attacks. Currently, specific indicators of compromise (e.g. IP addresses, domains, hashsums of malicious files) are shared in a semi-automated and structured way via so-called threat feeds. Unfortunately, current systems have to deal with the trade-off between the timeliness of such an alert (i.e. warning at the first mention of a threat) and the need to wait for verification by other sources (i.e. warning after multiple sources have verified the threat). In addition, due to the increasing number of open sources, it is challenging to find the right balance between feasibility and costs in order to identify a relatively small subset of valuable sources. In this paper, a method to automate the assessment of cyber threat intelligence sources and predict a relevance score for each source is proposed. Specifically, a model based on meta-data and word embedding is defined and experimented by training regression models to predict the relevance score of sources on Twitter. The results evaluation show that the assigned score allows to reduce the waiting time for intelligence verification, on the basis of its relevance, thus improving the time advantage of early threat detection.

      January 2022  Behaviour & Information Technology Article

    Nudging Users Towards Better Security Decisions in Password Creation Using Whitebox-based Multidimensional Visualizations

    Katrin Hartwig, Christian Reuter

    BibTeX DOI: 10.1080/0144929X.2021.1876167

    Abstract
    Nudging users to keep them secure online has become a growing research field in cybersecurity. While existing approaches are mainly blackbox based, showing aggregated visualisations as one-size-fits-all nudges, personalisation turned out promising to enhance the efficacy of nudges within the high variance of users and contexts. This article presents a disaggregated whitebox-based visualisation of critical information as a novel nudge. By segmenting users according to their decision-making and information processing styles, we investigate if the novel nudge is more effective for specific users than a common black-box nudge. Based on existing literature about critical factors in password security, we designed a dynamic radar chart and parallel coordinates as disaggregated visualisations. We evaluated the short-term effectiveness and users’ perception of the nudges in a think-aloud prestudy and a representative online evaluation (N=1.012). Our findings suggest that dynamic radar charts present a moderately effective nudge towards stronger passwords regarding short-term efficacy and are appreciated particularly by players of role-playing games.

      2022  Universitäts- und Landesbibliothek Darmstadt Darmstadt Other

    Spezialbericht - Smart-City-Ranking 2022 - Ressourcenschonend und CO₂-neutral - Die smarte Transformation unserer Städte im Hinblick auf die Energiekrise

    Arno Haselhorst, Jürgen Germies, Lucía Wright-Contreras, Luiza Camara, Annette Rudolph-Cleff, Joachim Schulze

    PDF BibTeX DOI: 10.26083/tuprints-00022871

    Abstract
    Welche Möglichkeiten stehen Städten zur Verfügung, um mit Hilfe der Digitalisierung den CO₂-Ausstoß zu reduzieren? Antworten auf diese Frage liefert ein Spezialbericht, welcher in der Zusammenarbeit zwischen Haselhorst Associates GmbH und der Technischen Universität Darmstadt entstanden ist. Unter dem Titel “Ressourcenschonend und CO₂-neutral: Die digitale Transformation unserer Städte im Hinblick auf die Energiekrise” widmet sich dieser Bericht dem Zusammenhang zwischen einer smarten und nachhaltigen Stadtentwicklung und zeigt auf, welche Handlungsmaßnahmen Städte und Stadtwerke angesichts der Energiekrise jetzt ergreifen sollten.

      2022 Conference

    Conference Proceedings : Science, Peace, Security ‘21 : The Impact of new Technologies: Destabilizing or Enabling Resilience? : 8-10 September 2021

    Malte Göttsche, Sibylle Bauer, Anja Dahlmann, Friederike Frieß, Filippa Lentzos, Götz Neuneck, Irmgard Niemeyer, Thea Riebe, Jantje Silomon, Christian Reuter, Jakob Brochhaus, Lukas Rademacher

    BibTeX DOI: 10.18154/RWTH-2022-02256

    Abstract
    The impact of the rapid technological change on peace and security continuously grows and becomes increasingly complex. Against the background of a quickly deteriorating security environment, the international conference SCIENCE · PEACE · SECURITY ’21 (RWTH Aachen University, 8-10 September 2021) examined the role of emerging technologies. The 60 speakers and 220 participants came from the natural, technical and social sciences. Diplomats and representatives from international organisations participated in the discussions. Topics included nuclear, chemical and biological arms control, autonomy in weapon systems, cybersecurity and the militarization of space among others. The main outcome was that all of these issues could be more effectively addressed by new approaches to rigorous interdisciplinary research collaboration to create policy-relevant knowledge and by tightening the nexus between the scientist and policy communities. Both can only be achieved and sustained by funding novel structures that enable scientific-technical scholars to engage on these topics. Key problems to be addressed by integrating natural, technical and social science perspectives include early risk assessment of potential dual-use research and technologies – especially in bio-security and epidemiology as well as IT and robotic research. Ways forward are the inclusion of norms into technology design as well as addressing questions of responsibility and standards. For military-usable technologies, entirely new regulatory approaches are necessary to prevent escalatory dynamics and to maintain accountability structures, moving from object-based to behaviour-based approaches. Scientific-technical research contributes to peace and security in positive ways. A prominent example are nuclear verification techniques. While instruments to monitor nonproliferation and test ban commitments benefit from further improvement, many gaps on how to verify future arms control and disarmament agreements still exist and must be urgently closed. The best cutting-edge scientific and academic expertise that is required for these complex research tasks is found in universities and other independent research institutes. Typically, however, decisionmakers draw knowledge from governmental institutions because of ease and existing connections. Therefore, efforts should be made to better connect the policy and academic communities. Communication between both can be improved by meeting on a regular basis and not only when advice on a specific issue is sought. This can foster more stable relationships and increase an understanding of each other. Lastly, opportunities should be improved for the younger generation of scientists and technologists to engage with policymakers. It is crucial to educate and engage early-on the next generation of scientifically-literate policymakers and security-aware scientists.

      2022  Technische Universität Darmstadt Darmstadt Thesis

    Algorithmisch gestützte Planung dezentraler Fluidsysteme

    Tim Moritz Müller

    BibTeX

    Abstract
    Fluidsysteme, wie Kühlkreisläufe oder die Wasserversorgung, sind essenziell für Industrie und Gesellschaft. Aufgrund ihres hohen Energieverbrauchs, ca. 1/3 des weltweiten Strombedarfs, sind jedoch Maßnahmen zur Reduktion der benötigten Eingangsleistung notwendig. Anhand der in dieser Arbeit betrachteten Pumpensysteme zeigen sich zwei Wege, dies zu realisieren: Zum einen kann die benötigte hydraulische Leistung durch dezentrale, verteilte Pumpen gesenkt werden. Zum anderen kann der Systemwirkungsgrad erhöht werden, wofür die Komponenten bestmöglich aufeinander abgestimmt werden müssen. Aus der sich hieraus ergebenden Komplexität folgen bei konventioneller Planung häufig subjektive, intransparente und vor allem suboptimale Entscheidungen. In der Arbeit wird sowohl das Potenzial für Kosten- und Energieeinsparungen durch Dezentralisierung, also auch die algorithmisch gestützte Planung zu Beherrschung der Komplexität behandelt. Dazu wird das Planungsproblem als gemischt-ganzzahliges, nichtlineares Optimierungsproblem formuliert und mittels problemspezifischer, algorithmisch gestützter Methoden gelöst. Als Anwendungsfälle werden Druckerhöhungsanlagen für die Wasserversorgung in Gebäuden sowie ein industrieller Kühlkreislauf betrachtet. Die Ergebnisse werden aus techno-ökonomischer Sicht diskutiert und experimentell validiert. In den Anwendungen können hohe Energie- und Kosteneinsparungen durch Dezentralisierung aufgezeigt werden. Aufgrund der erhöhten Komplexität ist hierzu jedoch eine algorithmisch gestützte Planung notwendig. Es wird zudem gezeigt, dass der Entscheidungsfindungsprozess transparenter gestaltet werden kann, indem Zielkonflikte mittels Pareto-Fronten klar dargelegt werden und energetische Schranken zum Aufzeigen des Potenzials genutzt werden.

      2022  Mensch und Computer 2022: Facing Realities Conference

    Detecting a Crisis: Comparison of Self-Reported vs. Automated Internet Outage Measuring Methods

    Denis Orlov, Simon Möller, Sven Düfer, Steffen Haesler, Christian Reuter

    PDF BibTeX DOI: 10.18420/muc2022-mci-ws10-321

    Abstract
    Every day, there are internet disruptions or outages around the world that affect our daily lives. In this paper, we analyzed these events in Germany in recent years and found out how they can be detected, and what impact they have on citizens, especially in crisis situations. For this purpose, we take a look at two different approaches to recording internet outages, namely the self-reporting of citizens and automatic reporting by algorithmic examination of the availability of IP networks. We evaluate the data of six major events with regard to their meaningfulness in quality and quantity. We found that due to the amount of data and the inherent imprecision of the methods used, it is difficult to detect outages through algorithmic examination. But once an event is publicly known by self-reporting, they have advantages to capture the temporal and spatial dimensions of the outage due to its nature of objective measurements. As a result, we propose that users’ crowdsourcing can enhance the detection of outages and should be seen as an important starting point to even begin an analysis with algorithm-based techniques, but it is to ISPs and regulatory authorities to support that.

      2022  2nd Workshop on Mobile Resilience: Designing Interactive Systems for Crisis Response Conference

    Proceedings of the 2nd Workshop on Mobile Resilience: Designing Interactive Systems for Crisis Response

    PDF BibTeX DOI: 10.26083/tuprints-00020092

    Abstract
    Information and communication technologies (ICT), including artificial intelligence, internet of things, and mobile applications can be utilized to tackle important societal challenges, such as the ongoing COVID-19 pandemic. While they may increase societal resilience, their design, functionality, and underlying infrastructures must be resilient against disruptions caused by anthropogenic, natural and hybrid crises, emergencies, and threats. In order to research challenges, designs, and potentials of interactive technologies, this workshop investigated the space of mobile technologies and resilient systems for crisis response, including the application domains of cyber threat and pandemic response.

      2022  Proceedings on Privacy Enhancing Technologies Conference

    On the Challenges of Developing a Concise Questionnaire to Identify Privacy Personas

    Tom Biselli, Enno Steinbrink, Franziska Herbert, Gina M. Schmidbauer-Wolf, Christian Reuter

    BibTeX DOI: 10.56553/popets-2022-0126

    Abstract
    Concise instruments to determine privacy personas – typical privacy-related user groups – are not available at present. Consequently, we aimed to identify them on a privacy knowledge–privacy behavior ratio based on a self-developed instrument. To achieve this, we conducted an item analysis (N = 820) and a confirmatory factor analysis (CFA) (N = 656) of data based on an online study with German participants. Starting with 81 items, we reduced those to an eleven-item questionnaire with the two scales privacy knowledge and privacy behavior. A subsequent cluster analysis (N = 656) revealed three distinct user groups: (1) Fundamentalists scoring high in privacy knowledge and behavior, (2) Pragmatists scoring average in privacy knowledge and behavior and (3) Unconcerned scoring low in privacy knowledge and behavior. In a closer inspection of the questionnaire, the CFAs supported the model with a close global fit based on RMSEA in a training and to a lesser extent in a cross-validation sample. Deficient local fit as well as validity and reliability coefficients well below generally accepted thresholds, however, revealed that the questionnaire in its current form cannot be considered a suitable measurement instrument for determining privacy personas. The results are discussed in terms of related persona conceptualizations, the importance of a methodologically sound investigation of corresponding privacy dimensions and our lessons learned.

      2022  DeSIRE Conference 2022 Conference

    Urban data platforms and urban critical infrastructure

    Michaela Leštáková, Frank Hessel, Kevin Logan, Yasin Alhamwy, Andreas Morgen, Martin Pietsch

    BibTeX

    Abstract
    Urban data platforms (UDP) are currently being launched in many cities as a part of their smart city strategies. They gather and provide access to data from various urban domains, including critical infrastructure. We performed a survey about UDPs in Germany. Focusing on their potential for improving resilience of the city (resilience through ICT) and the resilience of the UDPs themselves (resilience for ICT), our key findings were: ▪ UDP providers tend to focus on normal conditions rather than crisis ▪ critical infrastructure is often not covered ▪ lack of focus on crisis shows in the design of the UDPs as well

      2022  35th European Conference on Object-Oriented Programming Conference

    Prisma: A tierless language for enforcing contract-client protocols in decentralized apps

    David Richter, David Kretzler, Pascal Weisenburger, Guido Salvaneschi, Sebastian Faust, Mira Mezini

    PDF BibTeX DOI: 10.4230/LIPIcs.ECOOP.2022.35

    Abstract
    Decentralized applications (dApps) consist of smart contracts that run on blockchains and clients that model collaborating parties. dApps are used to model financial and legal business functionality. Today, contracts and clients are written as separate programs – in different programming languages – communicating via send and receive operations. This makes distributed program flow awkward to express and reason about, increasing the potential for mismatches in the client-contract interface, which can be exploited by malicious clients, potentially leading to huge financial losses. In this paper, we present Prisma, a language for tierless decentralized applications, where the contract and its clients are defined in one unit and pairs of send and receive actions that “belong together” are encapsulated into a single direct–style operation, which is executed differently by sending and receiving parties. This enables expressing distributed program flow via standard control flow and renders mismatching communication impossible. We prove formally that our compiler preserves program behavior in presence of an attacker controlling the client code. We systematically compare Prisma with mainstream and advanced programming models for dApps and provide empirical evidence for its expressiveness and performance.

      2022  Behaviour & Information Technology Article

    A quarter century of usable security and privacy research: transparency, tailorability, and the road ahead

    Christian Reuter, Luigi Lo Iacono, Alexander Benlian

    PDF BibTeX DOI: 10.1080/0144929X.2022.2080908

    Abstract
    In the last decades, research has shown that both technical solutions and user perceptions are important to improve security and privacy in the digital realm. The field of ‘usable security’ already started to emerge in the mid-90s, primarily focussed on password and email security. Later on, the research field of ”usable security and privacy” evolved and broadened the aim to design concepts and tools to assist users in enhancing their behaviour with regard to both privacy and security. Nevertheless, many user interventions are not as effective as desired. Because of highly diverse usage contexts, leading to different privacy and security requirements and not always to one-size-fits-all approaches, tailorability is necessary to address this issue. Furthermore, transparency is a crucial requirement, as providing comprehensible information may counter reactance towards security interventions. This article first provides a brief history of the research field in its first quarter-century and then highlights research on the transparency and tailorability of user interventions. Based on this, this article then presents six contributions with regard to (1) privacy concerns in times of COVID-19, (2) authentication on mobile devices, (3) GDPR-compliant data management, (4) privacy notices on websites, (5) data disclosure scenarios in agriculture, as well as (6) rights under data protection law and the concrete process should data subjects want to claim those rights. This article concludes with several research directions on user-centred transparency and tailorability.

      2022  Sensors Article

    Improving Wearable-Based Activity Recognition Using Image Representations

    Alejandro Sanchez Guinea, Mehran Sarabchian, Max Mühlhäuser

    PDF BibTeX DOI: 10.3390/s22051840

    Abstract
    Activity recognition based on inertial sensors is an essential task in mobile and ubiquitous computing. To date, the best performing approaches in this task are based on deep learning models. Although the performance of the approaches has been increasingly improving, a number of issues still remain. Specifically, in this paper we focus on the issue of the dependence of today’s state-of-the-art approaches to complex ad hoc deep learning convolutional neural networks (CNNs), recurrent neural networks (RNNs), or a combination of both, which require specialized knowledge and considerable effort for their construction and optimal tuning. To address this issue, in this paper we propose an approach that automatically transforms the inertial sensors time-series data into images that represent in pixel form patterns found over time, allowing even a simple CNN to outperform complex ad hoc deep learning models that combine RNNs and CNNs for activity recognition. We conducted an extensive evaluation considering seven benchmark datasets that are among the most relevant in activity recognition. Our results demonstrate that our approach is able to outperform the state of the art in all cases, based on image representations that are generated through a process that is easy to implement, modify, and extend further, without the need of developing complex deep learning models.

      2022  DeSIRE Conference 2022 Conference

    Multi-agent control of fluid systems – comparison of approaches

    Kevin Logan, Marius Stürmer, Tim Müller, Tobias Meck, Peter F. Pelz

    BibTeX

    Abstract
    Multi-agent systems allow system-wide, energy efficient control of fluid systems, fulfilling volume flow demands even in the face of disruptions within the communication network.

      December 2021  60th IEEE Conference on Decision and Control (CDC2021) Conference

    Discrete-Time Mean Field Control with Environment States

    Kai Cui, Anam Tahir, Mark Sinzger, Heinz Koeppl

    BibTeX DOI: 10.1109/CDC45484.2021.9683749

    Abstract
    Multi-agent reinforcement learning methods have shown remarkable potential in solving complex multi-agent problems but mostly lack theoretical guarantees. Recently, mean field control and mean field games have been established as a tractable solution for large-scale multi-agent problems with many agents. In this work, driven by a motivating scheduling problem, we consider a discrete-time mean field control model with common environment states. We rigorously establish approximate optimality as the number of agents grows in the finite agent case and find that a dynamic programming principle holds, resulting in the existence of an optimal stationary policy. As exact solutions are difficult in general due to the resulting continuous action space of the limiting mean field Markov decision process, we apply established deep reinforcement learning methods to solve the associated mean field control problem. The performance of the learned mean field control policy is compared to typical multi-agent reinforcement learning approaches and is found to converge to the mean field performance for sufficiently many agents, verifying the obtained theoretical results and reaching competitive solutions.

      December 2021  IEEE Global Communications Conference 2021 Conference

    Beamforming and link activation methods for energy efficient RIS-aided transmissions in C-RANs

    Jaime Quispe, Tarcisio Ferreira Maciel, Yuri Carvalho Barbosa Silva, Anja Klein

    BibTeX DOI: 10.1109/GLOBECOM46510.2021.9685593

    Abstract
    This work studies the application of a reconfigurable intelligent surface (RIS) in a cloud radio access network (C-RAN) targeting the reduction of resource usage while providing adequate capacity. We investigate if an RIS can contribute to improve the trade-off between the downlink system spectral efficiency (SE) and energy consumption of a multi-base-station (BS) multi-user single-RIS setup by means of link activation, radiated power control, and operational power mode decisions that can benefit from RIS-enhanced radio channels. For this purpose, we optimize the activations jointly with BS and RIS beamforming for maximum energy efficiency (EE) under a centralized approach and subject to SE, power, fronthaul capacity, and RIS phase-shift constraints. The associated mixed-boolean non-linear problem is solved using monotonic and semidefinite relaxation methods integrated in a Branch-Reduce-and-Bound procedure. Simulations show that the RIS helps to increase the EE of a C-RAN w.r.t. its non-RIS-aided and fully-connected versions by 30% and 80%, respectively.

      December 2021  IEEE Global Communications Conference 2021 Conference

    Joint Beamforming and BS Selection for Energy-Efficient Communications via Aerial-RIS

    Jaime Quispe, Tarcisio Ferreira Maciel, Yuri Carvalho Barbosa Silva, Anja Klein

    BibTeX DOI: 10.1109/GCWkshps52748.2021.9681981

    Abstract
    Cooperative BS transmission via unmanned aerial vehicles (UAVs)-airborne reconfigurable intelligent surface (RIS), also known as aerial-RIS, is a promising solution for providing connectivity in emergency areas where network access is unavailable. The RIS requires low power in reflecting the impinging base station (BS) signals towards the direction of the user equipment (UE), and the cooperative transmission can provide a more stable connection that guarantees quality-of-service (QoS). In this work, we investigate the energy efficiency (EE) maximization of a multiple-BS single-UE single-aerial-RIS setup and the usefulness of cooperation to prevent outages. The BSs can be turned either on or off depending on their contribution to the EE, and the system is subject to QoS, power, capacity, and RIS specific constraints. We formulate a problem that jointly optimizes the selection of the BSs and the beamforming weights of BSs and RIS, and solve it with a Branch-Reduce-and-Bound (BRnB) algorithm that uses monotonic optimization and semidefinite relaxation steps. Simulation results for an illustrative setup show that the aerial-RIS increases the EE by 50% when doubling the number of its elements and cooperative aerial-RIS transmissions help to solve outages of single-BS cases.

      December 2021  94th Vehicular Technology Conference (VTC2021-Fall) Conference

    Energy-Optimal Short Packet Transmission for Time-Critical Control

    Kilian Kiekenap, Andrea Patricia Ortiz Jimenez, Anja Klein

    BibTeX DOI: 10.1109/VTC2021-Fall52928.2021.9625205

    Abstract
    In this paper, the transmission energy for reliable communications with short packets and low latency requirements, e.g. for control applications, is minimized. Since the dynamics of the agents determine the allowed latencies for receiving control inputs, the requirements on latency and allowable packet error rate are individual, depending on the machine type. We consider a centralized environment with a single controller transmitting control commands wireless to multiple agents with given latency requirements. Also, the channel conditions are individual for each agent. Therefore, the optimal time-frequency resource allocation is derived for continuous time-frequency resource allocation. Since the resource allocation in OFDM systems like 5G is discrete, an algorithm to select the allocation from a resource grid with different resolutions is proposed and shown to achieve solutions with less than 0.5 dB increase in energy consumption compared to the continuous results. With numerical evaluation, the benefit of a channel-state- and deadline-aware solution is shown for a resource grid based on the 5G frame structure. On average, the gain of the proposed algorithm to an allocation only balancing the number of resources for each agent, as far as the deadlines allow, is about 50% energy saving.

      December 2021  94th Vehicular Technology Conference (VTC2021-Fall) Conference

    Energy Consumption Optimization for UAV Assisted Private Blockchain-based IIoT Networks

    Xinhua Lin, Jing Zhang, Lin Xiang, Xiohu Ge

    BibTeX DOI: 10.1109/VTC2021-Fall52928.2021.9625316

    Abstract
    The blockchain is a promising technology to enhance the security and resilience of industrial Internet of Things (IIoT) networks. However, generating blockchain for the IIoT devices usually consumes excessive energy which may not be affordable for battery-powered IIoT devices. To address this problem, in this paper, we consider an unmanned aerial vehicle (UAV) assisted private blockchain-based IIoT system. Thereby, a UAV mounted with computing processor is deployed as a multi-access edge computing platform, which is responsible for collecting data from the IIoT devices, generating blocks based on the collected data, and broadcasting the blocks to the IIoT devices. To minimize the energy consumption of the UAV, joint optimization of the central processing unit (CPU) frequencies for data computation and block generation, the amount of offloaded IIoT data, the bandwidth allocation, and the trajectory of the UAV is formulated as a nonconvex optimization problem and solved via a successive convex approximation (SCA) algorithm. Simulation results show that, compared with several baseline schemes, the proposed scheme can significantly lower the energy consumption required for the blockchain generation in IIoT networks.

      December 2021  29th European Signal Processing Conference (EUSIPCO) Conference

    Robust Spectral Clustering: A Locality Preserving Feature Mapping Based on M-estimation

    A. Taştan, M. Muma, A. M. Zoubir

    BibTeX DOI: 10.23919/EUSIPCO54536.2021.9616292

    Abstract
    Dimension reduction is a fundamental task in spectral clustering. In practical applications, the data may be corrupted by outliers and noise, which can obscure the underlying data structure. The effect is that the embeddings no longer represent the true cluster structure. We therefore propose a new robust spectral clustering algorithm that maps each high-dimensional feature vector onto a low-dimensional vector space. Robustness is achieved by posing the locality preserving feature mapping problem in form of a ridge regression task that is solved with a penalized M-estimation approach. An unsupervised penalty parameter selection strategy is proposed using the Fiedler vector, which is the eigenvector associated with the second smallest eigenvalue of a connected graph. More precisely, the penalty parameter is selected, such that, the corresponding Fiedler vector is Δ-separated with a minimum information loss on the embeddings. The method is benchmarked against popular embedding and spectral clustering approaches using real-world datasets that are corrupted by outliers.

      December 2021  Computers & Electrical Engineering Article

    A social media-based over layer on the edge for handling emergency-related events

    Andrea Tundis, Maksim Melnik, Hashim Naveed, Max Mühlhäuser

    PDF BibTeX DOI: 10.1016/j.compeleceng.2021.107570

    Abstract
    Online Social Networks (OSNs), together with messaging services are tools for the exchange of entertainment-related information. However, they represent virtual environments capable of providing relevant information related to emergency or criminal events. Thanks to the simple way of using OSNs in combination to modern ubiquitous Internet of Things (IoT) smart devices, the generation and exploitation of such information is made available to users in real-time even more easily. Unfortunately, its reuse has not been taken into consideration yet due to the lack of specific models and related software tools. In this context, the paper presents a social media-based over layer for supporting the monitoring, detection, computation and information sharing of social media information related to emergency scenarios centered on smartphones and text mining techniques. The proposal is assessed through two different case studies, by evaluating the performances of different classifiers and by showing the logic of the functionalities of the related apps.

      December 2021  Computers & Electrical Engineering Article

    An IoT-based context-aware model for danger situations detection

    Andrea Tundis, Muhammad Uzair, Max Mühlhäuser

    PDF BibTeX DOI: 10.1016/j.compeleceng.2021.107571

    Abstract
    On a daily basis, people perform planned or routine activities related to their needs, such as going to the office, playing sports and so on. Alongside them, unpleasant unforeseen situations can take place such as being robbed on the street or even being taken hostage. Providing information related to the crime scene or requesting help from the competent authorities is difficult. That is why, mechanisms to support users in such situations, based on their physical status, would be of great importance. Based on such idea, a context-aware model for detecting specific situations of danger is proposed. It is characterized by a set of defined features related to the body posture, the stress level and geolocation whose values are gathered through a smartphone and a smartwatch, as enabling technologies for the local computation. A machine learning technique was adopted for supporting body posture recognition, whereas a threshold-based approach was used to detect the stress level and to evaluate of user�s location. After the description of the proposed model, the achieved results as well as current limits are also discussed.

      November 2021  Transactions on Cryptographic Hardware and Embedded Systems Article

    Can’t Touch This: Inertial HSMs Thwart Advanced Physical Attacks

    Jan Sebastian Götte, Björn Scheuermann

    PDF BibTeX DOI: 10.46586/tches.v2022.i1.69-93

    Abstract
    In this paper, we introduce a novel countermeasure against physical attacks: Inertial Hardware Security Modules (IHSMs). Conventional systems have in common that their security requires the crafting of fine sensor structures that respond to minute manipulations of the monitored security boundary or volume. Our approach is novel in that we reduce the sensitivity requirement of security meshes and other sensors and increase the complexity of any manipulations by rotating the security mesh or sensor at high speed—thereby presenting a moving target to an attacker. Attempts to stop the rotation are easily monitored with commercial MEMS accelerometers and gyroscopes. Our approach leads to an HSM that can easily be built from off-the-shelf parts by any university electronics lab, yet offers a level of security that is comparable to commercial HSMs. We have built a proof-of-concept hardware prototype that demonstrates solutions to the concept’s main engineering challenges. As part of this proof-of-concept, we have found that a system using a coarse security mesh made from commercial printed circuit boards and an automotive high-g-force accelerometer already provides a useful level of security.

      November 2021  Conference on Empirical Methods in Natural Language Processing (EMNLP 2021) Conference

    MAD-G: Multilingual Adapter Generation for Efficient Cross-Lingual Transfer

    Alan Ansell, Maria Ponti Edoardo, Jonas Pfeiffer, Sebastian Ruder, Goran Glavaš, Ivan Vulić, Anna Korhonen

    PDF BibTeX

    Abstract
    Adapter modules have emerged as a general parameter-efficient means to specialize a pretrained encoder to new domains. Massively multilingual transformers (MMTs) have particularly benefited from additional training of language-specific adapters. However, this approach is not viable for the vast majority of languages, due to limitations in their corpus size or compute budgets. In this work, we propose MAD-G (Multilingual ADapter Generation), which contextually generates language adapters from language representations based on typological features. In contrast to prior work, our time- and space-efficient MAD-G approach enables (1) sharing of linguistic knowledge across languages and (2) zero-shot inference by generating language adapters for unseen languages. We thoroughly evaluate MAD-G in zero-shot cross-lingual transfer on part-of-speech tagging, dependency parsing, and named entity recognition. While offering (1) improved fine-tuning efficiency (by a factor of around 50 in our experiments), (2) a smaller parameter budget, and (3) increased language coverage, MAD-G remains competitive with more expensive methods for language-specific adapter training across the board. Moreover, it offers substantial benefits for low-resource languages, particularly on the NER task in low-resource African languages. Finally, we demonstrate that MAD-G’s transfer performance can be further improved via: (i) multi-source training, i.e., by generating and combining adapters of multiple languages with available task-specific training data; and (ii) by further fine-tuning generated MAD-G adapters for languages with monolingual data.

      October 2021   Jahrestagung der Gesellschaft für Informatik (INFORMATIK 2021) Conference

    Energy-efficient Mobile Sensor Data Offloading via WiFi using LoRa-based Connectivity Estimations

    Julian Zobel, Paul Frommelt, Patrick Lieser, Jonas Höchst, Patrick Lampe, Bernd Freisleben, Ralf Steinmetz

    BibTeX DOI: 10.18420/informatik2021-037

    Abstract
    Animal monitoring in natural habitats provides significant insights into the animals’ behavior, interactions, health, or external influences. However, the sizes of monitoring devices attachable to animals strongly depends on the animals’ sizes, and thus the range of possible sensors including batteries is severely limited. Gathered data can be offloaded from monitoring devices to data sinks in a wireless sensor network using available radio access technologies, but this process also needs to be as energy-efficient as possible. This paper presents an approach to combine the benefits of high-throughput WiFi and robust low-power LoRa communication for energy-efficient data offloading. WiFi is only used when connectivity between mobile devices and data sinks is available, which is determined by LoRa-based distance estimations without the need for additional GPS sensors. A prototypical implementation on low-end commodity-off-the-shelf hardware is used to evaluate the proposed approach in a German mixed forest using a simple path loss model for distance estimation. The system provides an offloading success rate of 87%, which is similar to that of a GPS-based approach, but with around 37% less power consumption.

      October 2021  5th European Conference on Mobile Robots Conference

    Industrial Manometer Detection and Reading for Autonomous Inspection Robots

    Jonas Günther, Martin Oehler, Stefan Kohlbrecher, Oskar von Stryk

    BibTeX DOI: 10.1109/ECMR50962.2021.9568833

    Abstract
    Autonomous mobile robots for industrial inspection can reduce cost for digitalization of existing plants by performing autonomous routine inspections. A frequent task is reading of analog gauges to monitor the health of the facility. Automating this process involves capturing image data with a camera sensor and processing the data to read the value. Detection algorithms deployed on a mobile robot have to deal with increased uncertainty regarding localization and environmental influences. This imposes increased requirements regarding robustness to viewing angle, lighting and scale variation on detection and reading. Current approaches based on conventional computer vision require high quality images or prior knowledge. We address these limitations by leveraging the advances of neural networks in the task of object detection and instance segmentation in a two-stage pipeline. Our method robustly detects and reads manometers without prior knowledge of object location or exact object type. In our evaluation we show that our approach can detect and read manometers from a distance of up to 3 m and a viewing angle of up to 60° in different lighting conditions with needle angle estimation errors of ±2.2°. We publish the validation split of our training dataset for manometer and needle detection at https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/2881.

      October 2021  5th European Conference on Mobile Robots Conference

    A Flexible Framework for Virtual Omnidirectional Vision to Improve Operator Situation Awareness

    Martin Oehler, Oskar von Stryk

    BibTeX DOI: 10.1109/ECMR50962.2021.9568840

    Abstract
    During teleoperation of a mobile robot, providing good operator situation awareness is a major concern as a single mistake can lead to mission failure. Camera streams are widely used for teleoperation but offer limited field-of-view. In this paper, we present a flexible framework for virtual projections to increase situation awareness based on a novel method to fuse multiple cameras mounted anywhere on the robot. Moreover, we propose a complementary approach to improve scene understanding by fusing camera images and geometric 3D Lidar data to obtain a colorized point cloud. The implementation on a compact omnidirectional camera reduces system complexity considerably and solves multiple use-cases on a much smaller footprint compared to traditional approaches such as actuated pan-tilt units. Finally, we demonstrate the generality of the approach by application to the multi-camera system of the Boston Dynamics Spot. The software implementation is available as open-source ROS packages on the project page https://tu-darmstadt-ros-pkg.github.io/omnidirectionalvision.

      October 2021  5th European Conference on Mobile Robots Conference

    Open-Source Tools for Efficient ROS and ROS2-based 2D Human-Robot Interface Development

    Stefan Fabian, Oskar von Stryk

    BibTeX DOI: 10.1109/ECMR50962.2021.9568801

    Abstract
    2D human-robot interfaces (HRI) are a key component of most robotic systems with an (optional) teleoperation component. However, creating such an interface is often cumbersome and time-consuming since most user interface frameworks require recompilation on each change or the writing of extensive boilerplate code even for simple interfaces. In this paper, we introduce five open-source packages, namely, the ros(2)babelfish packages, the qmlros(2)plugin packages, and the hectorrvizoverlay package. These packages enable the creation of visually appealing end-user or functionality-oriented diagnostic interfaces for ROS- and ROS2-based robots in a simple and quick fashion using the QtWidget or QML user interface framework. Optionally, rendering the interface as an overlay of the 3D scene of the robotics visualization tool rviz enables developers to leverage existing extensive data visualization capabilities.

      October 2021 Other

    The Terminating-Knockoff Filter: Fast High-Dimensional Variable Selection with False Discovery Rate Control

    J. Machkour, M. Muma, D. P. Palomar

    PDF BibTeX DOI: 10.48550/arXiv.2110.06048

    Abstract
    We propose the Terminating-Knockoff (T-Knock) filter, a fast variable selection method for high-dimensional data. The T-Knock filter controls a user-defined target false discovery rate (FDR) while maximizing the number of selected true positives. This is achieved by fusing the solutions of multiple early terminated random experiments. The experiments are conducted on a combination of the original data and multiple sets of randomly generated knockoff variables. A finite sample proof based on martingale theory for the FDR control property is provided. Numerical simulations show that the FDR is controlled at the target level while allowing for a high power. We prove under mild conditions that the knockoffs can be sampled from any univariate distribution. The computational complexity of the proposed method is derived and it is demonstrated via numerical simulations that the sequential computation time is multiple orders of magnitude lower than that of the strongest benchmark methods in sparse high-dimensional settings. The T-Knock filter outperforms state-of-the-art methods for FDR control on a simulated genome-wide association study (GWAS), while its computation time is more than two orders of magnitude lower than that of the strongest benchmark methods.

      October 2021  4th International Workshop on Emerging Technologies for Authorization and Authentication Conference

    Future-Proof Web Authentication: Bring Your Own FIDO2 Extensions

    Florentin Putz, Steffen Schön, Matthias Hollick

    PDF BibTeX DOI: 10.1007/978-3-030-93747-8_2

    Abstract
    The FIDO2 standards for strong authentication on the Internet define an extension interface, which allows them to flexibly adapt to future use cases. The domain of establishing new FIDO2 extensions, however, is currently limited to web browser developers and members of the FIDO alliance. We show how researchers and developers can design and implement their own extensions for using FIDO2 as a well-established and secure foundation to demonstrate innovative authentication concepts or to support custom deployments. Our open-source implementation targets the full FIDO2 stack, such as the Chromium web browser and hardware tokens, to enable tailor-made authentication based on the power of the existing FIDO2 ecosystem. To give an overview of existing extensions, we survey all published FIDO2 extensions by manually inspecting the source code of major web browsers and authenticators. Their current design, however, hinders the implementation of custom extensions, and they only support a limited number of extensions out of the box. We discuss weaknesses of current implementations and identify the lack of extension pass-through as a major limitation in current FIDO2 clients.

      October 2021  Wahrheit und Fake im postfaktisch-digitalen Zeitalter : Distinktionen in den Geistes- und IT-Wissenschaften Other

    Fake News technisch begegnen - Detektions- und Behandlungsansätze zur Unterstützung von NutzerInnen

    Katrin Hartwig, Christian Reuter

    PDF BibTeX DOI: 10.1007/978-3-658-32957-0_7

    Abstract
    Die Bedeutung des Umgangs mit Fake News hat sowohl im politischen als auch im sozialen Kontext zugenommen: Während sich bestehende Studien vor allem darauf konzentrieren, wie man gefälschte Nachrichten erkennt und kennzeichnet, fehlen Ansätze zur Unterstützung der NutzerInnen bei der eigenen Einschätzung weitgehend. Dieser Artikel stellt bestehende Black-Box- und White-Box-Ansätze vor und vergleicht Vor- und Nachteile. Dabei zeigen sich White-Box-Ansätze insbesondere als vielversprechend, um gegen Reaktanzen zu wirken, während Black-Box-Ansätze Fake News mit deutlich größerer Genauigkeit detektieren. Vorgestellt wird auch das von uns entwickelte Browser-Plugin TrustyTweet, welches die BenutzerInnen bei der Bewertung von Tweets auf Twitter unterstützt, indem es politisch neutrale und intuitive Warnungen anzeigt, ohne Reaktanz zu erzeugen.

      October 2021 Book

    Mastering Uncertainty in Mechanical Engineering

    Peter F. Pelz, Peter Groche, Marc E. Pfetsch, Maximilian Frederic Schäffner

    PDF BibTeX DOI: 10.1007/978-3-030-78354-9

    Abstract
    This open access book reports on innovative methods, technologies and strategies for mastering uncertainty in technical systems. Despite the fact that current research on uncertainty is mainly focusing on uncertainty quantification and analysis, this book gives emphasis to innovative ways to master uncertainty in engineering design, production and product usage alike. It gathers authoritative contributions by more than 30 scientists reporting on years of research in the areas of engineering, applied mathematics and law, thus offering a timely, comprehensive and multidisciplinary account of theories and methods for quantifying data, model and structural uncertainty, and of fundamental strategies for mastering uncertainty. It covers key concepts such as robustness, flexibility and resilience in detail. All the described methods, technologies and strategies have been validated with the help of three technical systems, i.e. the Modular Active Spring-Damper System, the Active Air Spring and the 3D Servo Press, which have been in turn developed and tested during more than ten years of cooperative research. Overall, this book offers a timely, practice-oriented reference guide to graduate students, researchers and professionals dealing with uncertainty in the broad field of mechanical engineering.

      September 2021  43rd DAGM German Conference on Pattern Recognition 2021 Conference

    TxT: Crossmodal End-to-End Learning with Transformers

    Jan-Martin O. Steitz, Jonas Pfeiffer, Iryna Gurevych, Stefan Roth

    PDF BibTeX DOI: 10.1007/978-3-030-92659-5_26

    Abstract
    Reasoning over multiple modalities, e.g. in Visual Question Answering (VQA), requires an alignment of semantic concepts across domains. Despite the widespread success of end-to-end learning, today’s multimodal pipelines by and large leverage pre-extracted, fixed features from object detectors, typically Faster R-CNN, as representations of the visual world. The obvious downside is that the visual representation is not specifically tuned to the multimodal task at hand. At the same time, while transformer-based object detectors have gained popularity, they have not been employed in today’s multimodal pipelines. We address both shortcomings with TxT, a transformer-based crossmodal pipeline that enables fine-tuning both language and visual components on the downstream task in a fully end-to-end manner. We overcome existing limitations of transformer-based detectors for multimodal reasoning regarding the integration of global context and their scalability. Our transformer-based multimodal model achieves considerable gains from end-to-end learning for multimodal question answering.

      September 2021  i-com: Journal of Interactive Media Article

    Towards Resilient Critical Infrastructures - Motivating Users to Contribute to Smart Grid Resilience

    Rolf Egert, Nina Gerber, Jasmin Haunschild, Philipp Kuehn, Verena Zimmermann

    BibTeX DOI: 10.1515/icom-2021-0021

    Abstract
    Smart cities aim at improving efficiency while providing safety and security by merging conventional infrastructures with information and communication technology. One strategy for mitigating hazardous situations and improving the overall resilience of the system is to involve citizens. For instance, smart grids involve prosumers—capable of producing and consuming electricity—who can adjust their electricity profile dynamically (i. e., decrease or increase electricity consumption), or use their local production to supply electricity to the grid. This mitigates the impact of peak consumption periods on the grid and makes it easier for operators to control the grid. This involvement of prosumers is accompanied by numerous socio-technical challenges, including motivating citizens to contribute by adjusting their electricity consumption to the requirements of the energy grid. Towards this end, this work investigates motivational strategies and tools, including nudging, persuasive technologies, and incentives, that can be leveraged to increase the motivation of citizens. We discuss long-term and side effects and ethical and privacy considerations, before portraying bug bounty programs, gamification and apps as technologies and strategies to communicate the motivational strategies to citizens.

      August 2021  Water Article

    A Method for Modeling Urban Water Infrastructures Combining Geo-Referenced Data

    Imke-Sophie Rehm, John Friesen, Kevin Pouls, Christoph Busch, Hannes Taubenböck, Peter F. Pelz

    PDF BibTeX DOI: 10.3390/w13162299

    Abstract
    Water distribution networks are the backbone of any municipal water supply. Their task is to supply the population regardless of the respective demand. High resilience of these infrastructures is of great importance and has brought these infrastructures into the focus of science and politics. At the same time, the data collected is highly sensitive and often openly unavailable. Therefore, researchers have to rely on models that represent the topology of these infrastructures. In this work, a model is developed that allows the topology of an urban water infrastructure to be mapped using the example of Cologne, Germany by combining freely available data. On the one hand, spatial data on land use (local climate zones) are used to disaggregate the water demand within the city under consideration. On the other hand, the parallelism of water and urban transportation infrastructures is used to identify the topology of a network by applying optimization methods. These networks can be analyzed to identify vulnerable areas within urban structures.

      August 2021  30th USENIX Security Symposium Conference

    PrivateDrop: Practical Privacy-Preserving Authentication for Apple AirDrop

    Alexander Heinrich, Matthias Hollick, Thomas Schneider, Milan Stute, Christian Weinert

    PDF BibTeX

    Abstract
    Apple’s offline file-sharing service AirDrop is integrated into more than 1.5 billion end-user devices worldwide. We discovered two design flaws in the underlying protocol that allow attackers to learn the phone numbers and email addresses of both sender and receiver devices. As a remediation, we study the applicability of private set intersection (PSI) to mutual authentication, which is similar to contact discovery in mobile messengers. We propose a novel optimized PSI-based protocol called PrivateDrop that addresses the specific challenges of offline resource-constrained operation and integrates seamlessly into the current AirDrop protocol stack. Using our native PrivateDrop implementation for iOS and macOS, we experimentally demonstrate that PrivateDrop preserves AirDrop’s exemplary user experience with an authentication delay well below one second. We responsibly disclosed our findings to Apple and open-sourced our PrivateDrop implementation.

      July 2021 Other

    Krisenfest durch dunkle Zeiten - Wie resilient sind deutsche Großstädte gegenüber Stromausfällen?

    Alice Knauf, Michèle Knodt

    PDF BibTeX DOI: 10.5281/zenodo.5082350

    Abstract
    Das System unserer kritischen Infrastrukturen wird komplexer und krisenanfälliger. Menschliches oder technisches Versagen, Naturkatastrophen, Pandemien, Cyber- oder Terrorangriffe können auch in Deutschland zu einem überregionalen Stromausfall führen, der länger als 24 Stunden anhält. Städte stehen dann als untere Katastrophenschutzbehörden vor der großen Herausforderung auf dieses Szenario zu reagieren und bis zu seiner Bewältigung möglichst gut durch die Krise zu kommen. An der Technischen Universität Darmstadt wurden im Rahmen von emergenCITY die Maßnahmen der lokalen Katastrophenschutzämter deutscher kreisfreier Großstädte auf das Szenario untersucht. Es zeigt sich, dass sich die meisten Katastrophenschutzämter mit dem Szenario auseinandersetzen. Dabei stehen interne Vorbereitungen im Bereich der Ressourcenausstattung im Vordergrund. Die Zusammenarbeit des Katastrophenschutzamtes beschränkt sich jedoch in vielen Städten auf einen einmaligen Austausch mit wenigen weiteren lokalen Akteuren. Um zukünftig gegenüber dem Szenario besser gewappnet zu sein, zeigen wir sechs Handlungsoptionen für häufig auftretende Problemfelder auf: Den Umgang mit dem Szenario üben Auf eine angespannte Personalsituation einstellen Katastrophenschutzamt personell stärken Bevölkerung in ihrer Vielfalt wahrnehmen Katastrophenschutz als Querschnittsaufgabe stärken Kooperative Formate verstetigen und ausbauen

      June 2021  14th ACM Conference on Security and Privacy in Wireless and Mobile Networks Conference

    AirCollect: Efficiently Recovering Hashed Phone Numbers Leaked via Apple AirDrop

    Alexander Heinrich, Matthias Hollick, Thomas Schneider, Milan Stute, Christian Weinert

    PDF BibTeX DOI: 10.1145/3448300.3468252

    Abstract
    Apple’s file-sharing service AirDrop leaks phone numbers and email addresses by exchanging vulnerable hash values of the user’s own contact identifiers during the authentication handshake with nearby devices. In a paper presented at USENIX Security’21, we theoretically describe two attacks to exploit these vulnerabilities and propose “PrivateDrop” as a privacy-preserving drop-in replacement for Apple’s AirDrop protocol based on private set intersection. In this demo, we show how these vulnerabilities are efficiently exploitable via Wi-Fi and physical proximity to a target. Privacy and security implications include the possibility of conducting advanced spear phishing attacks or deploying multiple “collector” devices in order to build databases that map contact identifiers to specific locations. For our proof-of-concept, we leverage a custom rainbow table construction to reverse SHA-256 hashes of phone numbers in a matter of milliseconds. We discuss the trade-off between success rate and storage requirements of the rainbow table and, after following responsible disclosure with Apple, we publish our proof-of-concept implementation as “AirCollect” on GitHub.

      June 2021  14th ACM Conference on Security and Privacy in Wireless and Mobile Networks Conference

    OpenHaystack: A Framework for Tracking Personal Bluetooth Devices via Apple’s Massive Find My Network

    Alexander Heinrich, Milan Stute, Matthias Hollick

    BibTeX DOI: 10.1145/3448300.3468251

    Abstract
    OpenHaystack is an open-source framework for locating personal Bluetooth devices using Apple’s Find My Network. A user can integrate it into Bluetooth-capable devices, such as notebooks, or create custom tracking accessories that can be attached to personal items (key rings, backpacks, etc.). We provide firmware images for the Nordic nRF5 chips and the ESP32. We show that they consume little energy and run from a single coin cell for a year. Our macOS application can locate personal accessories. Finally, we make both application and firmware available on GitHub.

      May 2021  Technische Universität Darmstadt Wiesbaden Book

    Information Refinement Technologies for Crisis Informatics: User Expectations and Design Implications for Social Media and Mobile Apps

    Marc-André Kaufhold

    PDF BibTeX

    Abstract
    Marc-André Kaufhold explores user expectations and design implications for the utilization of new media in crisis management and response. He develops a novel framework for information refinement, which integrates the event, organisational, societal, and technological perspectives of crises. Therefore, he reviews the state of the art on crisis informatics and empirically examines the use, potentials and barriers of both social media and mobile apps. Based on these insights, he designs and evaluates ICT concepts and artifacts with the aim to overcome the issues of information overload and quality in large-scale crises, concluding with practical and theoretical implications for technology adaptation and design.

      May 2021 Book

    Sicherheitskritische Mensch-Computer-Interaktion : Interaktive Technologien und Soziale Medien im Krisen- und Sicherheitsmanagement

    PDF BibTeX DOI: 10.1007/978-3-658-32795-8

    Abstract
    Die zweite, aktualisierte Auflage dieses Lehr- und Fachbuchs gibt eine fundierte und praxisbezogene Einführung sowie einen Überblick über Grundlagen, Methoden und Anwendungen der Mensch-Computer-Interaktion im Kontext von Sicherheit, Notfällen, Krisen, Katastrophen, Krieg und Frieden. Dies adressierend werden interaktive, mobile, ubiquitäre und kooperative Technologien sowie soziale Medien vorgestellt. Hierbei finden klassische Themen wie benutzbare (IT-)Sicherheit, Industrie 4.0, Katastrophenschutz, Medizin und Automobil, aber auch Augmented Reality, Crowdsourcing, Shitstorm Management, Social Media Analytics und Cyberwar ihren Platz. Methodisch wird das Spektrum von Usable Safety bis Usable Security Engineering von Analyse über Design bis Evaluation abgedeckt. Das Buch eignet sich ebenso als Lehrbuch für Studierende wie als Handbuch für Wissenschaftler, Designer, Entwickler und Anwender.

      April 2021  IEEE Transactions on Mobile Computing Article

    Performance and Pitfalls of 60 GHz WLANs Based on Consumer-Grade Hardware

    Swetank Kumar Saha, Hany Assasa, Adrian Loch, Naveen Muralidhar Prakash, Roshan Shyamsunder, Shivang Aggarwal, Daniel Steinmetzer, Dimitrios Koutsonikolas, Joerg Widmer, Matthias Hollick

    PDF BibTeX DOI: 10.1109/TMC.2020.2967386

    Abstract
    Wireless networks operating in the 60 GHz band have the potential to provide very high throughput but face a number of challenges (e.g., high attenuation, beam training, and coping with mobility) which are widely accepted but often not well understood in practice. Understanding these challenges, and especially their actual impact on consumer-grade hardware is fundamental to fully exploit the high physical layer rates in the 60 GHz band. To this end, we perform an extensive measurement campaign using two commercial off-the-shelf 60 GHz routers in real-world environments. Our results allow us to revisit a range of issues and provide much deeper insights into the reasons for specific performance compared to prior work on performance characterization. Further, our study goes beyond basic link characterization and explores for the first time practical considerations such as coverage and access point deployment. While some of our observations are expected, we also obtain highly surprising insights that challenge the prevailing wisdom in the community. We derive the shortcomings of current commercial 60 GHz devices, and the fundamental problems that remain open on the way to fast and efficient 60 GHz networking.

      March 2021   VDI/VDE Mechatronik‐Tagung 2021 Conference

    Entwicklung eines autonomiefokussierten hochmobilen Bodenrobotersystems für den Katastrophenschutz

    Marius Schnaubelt, Tobias Ullrich, Moritz Torchalla, Jonas Diegelmann, Matthias Hoffmann, Oskar von Stryk

    BibTeX

    Abstract
    Mobile Rettungsroboter ermöglichen den menschlichen Bedienern die Bearbeitung von Aufgaben aus sicherer Entfernung in risikoreichen Umgebungen. Durch die unstrukturierte Umgebung der komplexen und vorab unbekannten Einsatzszenarien, verursacht die aktuell übliche Teleoperation der Robotersysteme eine hohe kognitive Belastung für den Roboteroperator, was schnell zur Ermüdung führt. Durch intelligente autonome Assistenzfunktionen können die Operatoren entlastet werden, wodurch die Wahrscheinlichkeit von Bedienfehlern reduziert und die Effizienz des Robotereinsatzes erhöht werden kann. Diese innovativen Assistenzfunktionen benötigen jedoch ein mechatronisches Design, dessen Anforderungen an Hard- und Software für ein effektives Gesamtsystem eng aufeinander abgestimmt und umgesetzt werden müssen. Die Entwicklung eines hochmobilen autonomiefokussierten Bodenroboters mit modularen Sensornutzlasten ermöglicht dem Operator ein umfassendes Situationsbewusstsein sowie Unterstützung bei Navigation und Manipulation. Die Evaluation des Gesamtsystems und von Einzelkomponenten analysiert die Erfüllung des Anforderungskatalogs und demonstriert so die Eignung für (semi-)autonome Rettungsrobotikeinsätze.

      2021  2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR) Conference

    HectorGrapher: Continuous-time Lidar SLAM with Multi-resolution Signed Distance Function Registration for Challenging Terrain

    Kevin Daun, Marius Schnaubelt, Stefan Kohlbrecher, Oskar von Stryk

    BibTeX

    Abstract
    For deployment in previously unknown, unstructured, and GPS-denied environments, autonomous mobile rescue robots need to localize themselves in such environments and create a map of it using a simultaneous localization and mapping (SLAM) approach. Continuous-time SLAM approaches represent the pose as a time-continuous estimate that provides high accuracy and allows correcting for distortions induced by motion during the scan capture. To enable robust and accurate real-time SLAM in challenging terrain, we propose HectorGrapher which enables accurate localization by continuous-time pose estimation and robust scan registration based on multi-resolution signed distance functions. We evaluate the method in multiple publicly available real-world datasets, as well as a data set from the RoboCup 2021 Rescue League, where we applied the proposed method to win the Best-in-Class “Exploration and Mapping” Award.

      2021 Other

    Transformation, Zirkulation, System of Systems : Für ein dynamisches Verständnis netzgebundener Infrastrukturen

    Jens Ivo Engels, Sybille Frank, Iryna Gurevych, Martina Heßler, Michèle Knodt, Jochen Monstadt, Alfred Nordmann, Andreas Oetting, Annette Rudolph-Cleff, Uwe Rüppel, Gerrit Jasper Schenk, Florian Steinke

    PDF BibTeX DOI: 10.26083/tuprints-00017923

    Abstract
    Der Aufsatz plädiert dafür, die Dynamik technischer netzgebundener Infrastrukturen mit einem dreifachen Ansatz zu untersuchen: Transformation, Zirkulation und System of Systems. Transformation repräsentiert dabei die Veränderung von Infrastrukturen als Gesamtsysteme. Zirkulation repräsentiert die in jeder Infrastrukturfunktion eingeschriebene Dynamik. Gemeint ist der Austausch von Gütern, Menschen, Informationen oder Energie in Netzen. System of Systems ist ein Konzept zur Beschreibung der Interdependenzen verschiedener verbundener Sektoren und Systeme. Die drei Merkmale der Transformation sind aufeinander bezogen: Das Konzept des System of System liefert einen Erklärungsansatz, auf welchen Wegen Zirkulation stattfindet (in und zwischen Netzwerken) und welche Faktoren die Zirkulation beeinflussen. Diese Phänomene können wiederum als Ursachen oder Anreize für Transformation auf der Systemebene begriffen werden. Umgekehrt ist zu fragen, inwieweit Transformationsprozesse der Infrastrukturnetze die Konfiguration des System of Systems verändern.

      2021  2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR) Conference

    Robust Multisensor Fusion for Reliable Mapping and Navigation in Degraded Visual Conditions

    Moritz Torchalla, Marius Schnaubelt, Kevin Daun, Oskar von Stryk

    BibTeX

    Abstract
    We address the problem of robust simultaneous mapping and localization in degraded visual conditions using low-cost off-the-shelf radars. Current methods often use high- end radar sensors or are tightly coupled to specific sensors, limiting the applicability to new robots. In contrast, we present a sensor-agnostic processing pipeline based on a novel forward sensor model to achieve accurate updates of signed distance function-based maps and robust optimization techniques to reach robust and accurate pose estimates. Our evaluation demonstrates accurate mapping and pose estimation in indoor environments under poor visual conditions and higher accuracy compared to existing methods on publicly available benchmark data.

      2021  35th International Conference on Neural Information Processing Systems (NeurIPS 2021) Conference

    Implicit Generative Copulas

    Tim Janke, Mohamed Ghanmi, Florian Steinke

    PDF BibTeX

    Abstract
    Copulas are a powerful tool for modeling multivariate distributions as they allow to separately estimate the univariate marginal distributions and the joint dependency structure. However, known parametric copulas offer limited flexibility especially in high dimensions, while commonly used non-parametric methods suffer from the curse of dimensionality. A popular remedy is to construct a tree-based hierarchy of conditional bivariate copulas.In this paper, we propose a flexible, yet conceptually simple alternative based on implicit generative neural networks.The key challenge is to ensure marginal uniformity of the estimated copula distribution.We achieve this by learning a multivariate latent distribution with unspecified marginals but the desired dependency structure.By applying the probability integral transform, we can then obtain samples from the high-dimensional copula distribution without relying on parametric assumptions or the need to find a suitable tree structure.Experiments on synthetic and real data from finance, physics, and image generation demonstrate the performance of this approach.

      2021  2021 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe) Conference

    Evaluation of Multiparametric Linear Programming for Economic Dispatch under Uncertainty

    Johannes Sindt, Allan Santos, Marc E. Pfetsch, Florian Steinke

    BibTeX

    Abstract
    For risk assessment purposes, we study how economic dispatch decisions vary with the uncertain input factors that may arise, e.g., from the use of variable renewable energies. Given a known random input distribution and linear programming (LP)-based dispatch, we aim to describe the distribution of the resulting variables and objective values. Relying on Monte Carlo simulation (MCS) is computationally expensive, especially if the uncertain factors are high dimensional. In this paper we evaluate an algorithm using multiparametric linear programming (MPLP) for this purpose. It avoids solving an LP for every sample of the random vector by characterizing the parametric LP solution as a piece-wise linear function whose pieces can be stored for repeated use. We compare the algorithm with MCS and other quasi-Monte Carlo sampling approaches for three economic dispatch use cases with varying complexity. The MPLP approach is as accurate as MCS, but up to 300 times faster for the merit order use case.

      2021  30th USENIX Security Symposium Conference

    Disrupting Continuity of Apple’s Wireless Ecosystem Security: New Tracking, DoS, and MitM Attacks on iOS and macOS Through Bluetooth Low Energy, AWDL, and Wi-Fi

    Milan Stute, Alexander Heinrich, Jannik Lorenz, Matthias Hollick

    PDF BibTeX

    Abstract
    Apple controls one of the largest mobile ecosystems, with 1.5 billion active devices worldwide, and offers twelve proprietary wireless Continuity services. Previous works have unveiled several security and privacy issues in the involved protocols. These works extensively studied AirDrop while the coverage of the remaining vast Continuity service space is still low. To facilitate the cumbersome reverse-engineering process, we describe the first guide on how to approach a structured analysis of the involved protocols using several vantage points available on macOS. Also, we develop a toolkit to automate parts of this otherwise manual process. Based on this guide, we analyze the full protocol stacks involved in three Continuity services, in particular, Handoff (HO), Universal Clipboard (UC), and Wi-Fi Password Sharing (PWS). We discover several vulnerabilities spanning from Bluetooth Low Energy (BLE) advertisements to Apple’s proprietary authentication protocols. These flaws allow for device tracking via HO’s mDNS responses, a denial-of-service (DoS) attack on HO and UC, a DoS attack on PWS that prevents Wi-Fi password entry, and a machine-in-the-middle (MitM) attack on PWS that connects a target to an attacker-controlled Wi-Fi network. Our PoC implementations demonstrate that the attacks can be mounted using affordable off-the-shelf hardware ($20 micro:bit and a Wi-Fi card). Finally, we suggest practical mitigations and share our findings with Apple, who have started to release fixes through iOS and macOS updates.

      2021  24th International Conference on Artificial Intelligence and Statistics Conference

    Approximately Solving Mean Field Games via Entropy-Regularized Deep Reinforcement Learning

    Kai Cui, Heinz Koeppl

    BibTeX

    Abstract
    The recent mean field game (MFG) formalism facilitates otherwise intractable computation of approximate Nash equilibria in many-agent settings. In this paper, we consider discrete-time finite MFGs subject to finite-horizon objectives. We show that all discrete-time finite MFGs with non-constant fixed point operators fail to be contractive as typically assumed in existing MFG literature, barring convergence via fixed point iteration. Instead, we incorporate entropy-regularization and Boltzmann policies into the fixed point iteration. As a result, we obtain provable convergence to approximate fixed points where existing methods fail, and reach the original goal of approximate Nash equilibria. All proposed methods are evaluated with respect to their exploitability, on both instructive examples with tractable exact solutions and high-dimensional problems where exact methods become intractable. In high-dimensional scenarios, we apply established deep reinforcement learning methods and empirically combine fictitious play with our approximations.

      December 2020  IEEE Radar Conference Conference

    An Unsupervised Approach for Graph-based Robust Clustering of Human Gait Signatures

    A. Taştan, M. Muma, A. M. Zoubir

    BibTeX DOI: 10.1109/RadarConf2043947.2020.9266313

    Abstract
    Classification of gait abnormalities plays a key role in medical diagnosis, sports, physiotherapy and rehabilitation. We demonstrate the effectiveness of a new graph construction-based outlier detection method and and the applicability of a new parameter-free clustering approach on radar-based human gait signatures. Micro-Doppler radar-based human gait signatures of ten test subjects for five different gait types consisting of normal, simulated abnormal and assisted walks are clustered using five different clustering algorithms. The proposed algorithm outperforms existing methods both in cluster enumeration and partition and achieves an overall correct clustering rate of 92.8%. The developed method is promising for performing medical diagnosis in a robust unsupervised fashion.

      November 2020  45th Local Computer Networks Symposium on Emerging Topics in Networking Conference

    Topology-aware Path Planning for In-Transit Coverage of Aerial Post-Disaster Communication Assistance Systems

    Julian Zobel, Benjamin Becker, Ralf Kundel, Patrick Lieser, Ralf Steinmetz

    BibTeX DOI: 10.1109/LCNSymposium50271.2020.9363268

    Abstract
    The increase in natural disasters that impair and destroy communication infrastructure over the last years simultaneously increased the importance of infrastructure-independent ad hoc communication. Especially delay-tolerant networks (DTNs) are able to provide basic communication functionality for civilians, but performance suffers from a typically highly intermittent network with clusters around important locations like shelters. Small Unmanned Aerial Vehicles (UAVs) have proven to be efficient data ferries between clusters, but they require knowledge of cluster locations and also do not cover network nodes in transit between clusters. These in-transit nodes are therefore disconnected from the network for a long time and might miss critical messages like evacuation notices or hazard warnings. This paper provides two contributions for UAV-assisted post-disaster DTN communication. First, we present a novel approach to estimate the location of dynamically changing clusters in a post-disaster scenario. Second, we introduced a topology-aware path planning approach for UAV data ferry flights, covering in-transit node on their way between clusters. Our evaluation results highlight the quality requirements on topology information for an efficient application of Aerial Post-Disaster Communication Assistance Systems and demonstrate the positive impact of in-transit node coverage on the DTN’s communication performance.

      November 2020  SPLASH ‘20: Conference on Systems, Programming, Languages, and Applications, Software for Humanity Conference

    ConSysT: Tunable, Safe Consistency Meets Object-Oriented Programming

    Mirko Köhler, Nafise Eskandani, Alessandro Margara, Guido Salvaneschi

    BibTeX DOI: 10.1145/3427761.3428346

    Abstract
    Data replication is essential in scenarios like geo-distributed datacenters, but poses challenges for data consistency. Developers adopt Strong consistency at the cost of performance or embrace Weak consistency and face a higher programming complexity. We argue that languages should associate consistency to data types. We present , a programming language and middleware that provides abstractions to specify consistency types, enabling mixing different consistency levels in the same application. Such mechanism is fully integrated with object-oriented programming and type system guarantees that different levels can only be mixed correctly.

      November 2020  Proceedings of the ACM on Programming Languages Article

    Rethinking Safe Consistency in Distributed Object-Oriented Programming

    Mirko Köhler, Nafise Eskandani, Pascal Weisenburger, Alessandro Margara, Guido Salvaneschi

    BibTeX DOI: 10.1145/3428256

    Abstract
    Large scale distributed systems require to embrace the trade off between consistency and availability, accepting lower levels of consistency to guarantee higher availability. Existing programming languages are, however, agnostic to this compromise, resulting in consistency guarantees that are the same for the whole application and are implicitly adopted from the middleware or hardcoded in configuration files. In this paper, we propose to integrate availability in the design of an object-oriented language, allowing developers to specify different consistency and isolation constraints in the same application at the granularity of single objects. We investigate how availability levels interact with object structure and define a type system that preserves correct program behavior. Our evaluation shows that our solution performs efficiently and improves the design of distributed applications.

      October 2020  Proceedings of the ACM on Human-Computer Interaction Article

    Countering Fake News: A Comparison of Possible Solutions Regarding User Acceptance and Effectiveness

    Jan Kirchner, Christian Reuter

    BibTeX DOI: 10.1145/3415211

    Abstract
    Since the emergence of so-called fake news on the internet and in social media, platforms such as Facebook have started to take countermeasures, and researchers have begun looking into this phenomenon from a variety of perspectives. A large number of scientific work has investigated ways to detect fake news automatically. Less attention has been paid to the subsequent step, i.e., what to do when you are aware of the inaccuracy of claims in social media. This work takes a user-centered approach on means to counter identified mis- and disinformation in social media. We conduct a three-step study design on how approaches in social media should be presented to respect the users’ needs and experiences and how effective they are. As our first step, in an online survey representative for some factors to the German adult population, we enquire regarding their strategies on handling information in social media, and their opinion regarding possible solutions — focusing on the approach of displaying a warning on inaccurate posts. In a second step, we present five potential approaches for countermeasures identified in related work to interviewees for qualitative input. We discuss (1) warning, (2) related articles, (3) reducing the size, (4) covering, and (5) requiring confirmation. Based on the interview feedback, as the third step of this study, we select, improve, and examine four promising approaches on how to counter misinformation. We conduct an online experiment to test their effectiveness on the perceived accuracy of false headlines and also ask for the users’ preferences. In this study, we find that users welcome warning-based approaches to counter fake news and are somewhat critical with less transparent methods. Moreover, users want social media platforms to explain why a post was marked as disputed. The results regarding effectiveness are similar: Warning-based approaches are shown to be effective in reducing the perceived accuracy of false headlines. Moreover, adding an explanation to the warning leads to the most significant results. In contrast, we could not find a significant effect on one of Facebook’s current approaches (reduced post size and fact-checks in related articles).

      October 2020  MobileHCI ‘20: 22nd International Conference on Human-Computer Interaction with Mobile Devices and Services Conference

    Crisis Volunteering Nerds: Three Months After COVID-19 Hackathon #WirVsVirus

    Steffen Haesler, Stefka Schmid, Christian Reuter

    BibTeX DOI: 10.1145/3406324.3424584

    Abstract
    The hackathon #WirVsVirus in March 2020 was one of the biggest hackathons in history. Under the patronage of the federal government of Germany, 28,361 participants worked together in 1,498 projects, finding innovative apps and solutions against the COVID-19 pandemic. Three months after the event, we present an exemplifying analysis of the topics, used technologies and remaining activity of these projects. Shedding light on this instance of citizen science allows to highlight the potential of hackathons and startup culture regarding socio-technological resilience. At the same time, it may be understood as an impulse for crisis informatics to consider new forms of volunteering in the course of crisis management.

      September 2020  14th International Workshop on Wireless Network Testbeds, Experimental evaluation & Characterization Conference

    Hardware-Accelerated Real-Time Stream Data Processing on Android with GNU Radio

    Bastian Bloessl, Lars Baumgärtner, Matthias Hollick

    PDF BibTeX DOI: 10.1145/3411276.3412184

    Abstract
    With the ever-increasing performance of smartphones and tablets, they become viable platforms for applications that were, in the past, only possible on desktops or laptops. In this paper, we study their applicability for real-time stream-data processing, which is particularly interesting for Software Defined Radio (SDR) applications, enabling wireless measurement and experimentation campaigns on mobile platforms. To this end, we port GNU Radio, a state-of-theart, open source, real-time stream-data processing framework, to Android and evaluate its performance. We show that it is possible to fully benefit from available accelerators, i.e., Single Instruction Multiple Data (SIMD) and the Graphics Processing Unit (GPU), which provide considerable speedups and allow for efficient implementations. As a general-purpose real-time data processing framework, GNU Radio can provide the base for a wide range of applications. To demonstrate its flexibility, we provide example applications that implement FM and Wireless LAN (WLAN). Our toolchain is published as open source software, thus serving as an enabler for highly mobile SDR applications.

      August 2020  Datenschutz und Datensicherheit (DuD) Article

    Datensicherheit von Corona-Apps nach der DSGVO

    Tim Grube, Alexander Heinrich, Jan-Philipp Stroscher, Sabrina Schomberg

    BibTeX DOI: 10.1007/s11623-020-1314-0

    Abstract
    Der Beitrag analysiert die Protokolle der Konsortien DP-3T und PEPP-PT aus technischer Perspektive und grenzt diese voneinander ab. Zudem wird die technische Ausgestaltung der Entwicklerschnittstelle (API) von Google und Apple dargestellt. Aufbauend darauf erfolgt eine rechtliche Beurteilung der sich aus Art. 5 Abs. 1 lit. f, 25, 32 DSGVO ergebenden und die Datensicherheit betreffenden Kriterien und deren konkrete Umsetzung in den Protokollen.

      July 2020  13th ACM Conference on Security and Privacy in Wireless and Mobile Networks Conference

    Acoustic Integrity Codes: Secure Device Pairing Using Short-Range Acoustic Communication

    Florentin Putz, Flor Álvarez, Jiska Classen

    PDF BibTeX DOI: 10.1145/3395351.3399420

    Abstract
    Secure Device Pairing (SDP) relies on an out-of-band channel to authenticate devices. This requires a common hardware interface, which limits the use of existing SDP systems. We propose to use short-range acoustic communication for the initial pairing. Audio hardware is commonly available on existing off-the-shelf devices and can be accessed from user space without requiring firmware or hardware modifications. We improve upon previous approaches by designing Acoustic Integrity Codes (AICs): a modulation scheme that provides message authentication on the acoustic physical layer. We analyze their security and demonstrate that we can defend against signal cancellation attacks by designing signals with low autocorrelation. Our system can detect overshadowing attacks using a ternary decision function with a threshold. In our evaluation of this SDP scheme’s security and robustness, we achieve a bit error ratio below 0.1% for a net bit rate of 100 bps with a signal-to-noise ratio (SNR) of 14 dB. Using our open-source proof-of-concept implementation on Android smartphones, we demonstrate pairing between different smartphone models.

      June 2020  International Journal of Mechanics and Control Article

    Optimization-Based Planning for Autonomous Traversal of Obstacles with Mobile Ground Robots

    Martin Oehler, Stefan Kohlbrecher, Oskar von Stryk

    BibTeX

    Abstract
    Mobile robotic platforms which are traversing unstructured environments with challenging uneven terrain are permanently endangered of falling over. Previous research on trajectory planning methods for the prevention of vehicle tip-over is mostly limited to basic mobility systems with only few degrees of freedom (DOF). This paper proposes a novel optimization-based planning approach that enables mobile robots to autonomously traverse obstacles and rough terrain more safely. A 3D world model as provided from external sensors like Lidar is used to compute a whole-body motion plan in advance by optimizing the trajectories of each joint. Active flipper tracks maximize ground contact for improved traction and, if available, manipulator arm joints are used to further improve stability metrics. Additional constraints prevent collisions with the environment and the robot itself. The presented approach makes only few assumptions about the robot’s configuration and is applicable to a wide range of wheeled or tracked platforms. This is demonstrated by experimental evaluation for two different robots in simulation and for one physical robot. In four different test scenarios it is shown, that the proposed approach effectively prevents vehicle tip-over during traversal of uneven ground.

      June 2020  International Journal of Disaster Risk Reduction (IJDRR) Article

    Emergency service staff and social media – A comparative empirical study of the attitude by Emergency Services staff in Europe in 2014 and 2017

    Christian Reuter, Marc-André Kaufhold, Fabian Spahr, Thomas Spielhofer, Anna Sophie Hahne

    PDF BibTeX DOI: 10.1016/j.ijdrr.2020.101516

    Abstract
    Finding a way to ensure an effective use of social media has become increasingly important to emergency services over the past decade. Despite all efforts to determine the utility of social media for emergency organisations, it is necessary to benefit from such institutions’ staffs’ opinions to establish effective use. To provide empirical evidence we present a comparison of two surveys, conducted across Europe with emergency services in 2014 and 2017 respectively, with a total of 1169 answers. The analysis shows that personal experience has an effect on how organisational usage of social media is perceived and how emergency service staff view the future use of social media. Furthermore, the use has increased. This article not only shows emergency services what their staff think about their social media usage but also discusses challenges and future directions for the design of systems that can be useful for further development of optimized organisational social media usage.

      May 2020  28th European Conference on Information Systems Conference

    Warning the Public: A Survey on Attitudes, Expectations and Use of Mobile Crisis Apps in Germany

    Marc-André Kaufhold, Jasmin Haunschild, Christian Reuter

    PDF BibTeX

    Abstract
    As part of information systems, the research field of crisis informatics increasingly investigates the potentials and limitations of mobile crisis apps, which constitute a relatively new public service for citizens and are specifically designed for the dissemination of disaster‐related information and communication between authorities, organizations and citizens. While existing crisis apps, such as KATWARN or NINA in Germany, focus on preparatory information and warning functionality, there is a need for apps and research on police-related functionality, such as information on cybercrime, fraud offences, or search for missing persons. Based on a workshop with civil protection (N=12) and police officers (N=15), we designed a questionnaire and conducted a representative survey of German citizens (N=1.219) on the past, current and future use, perceived helpfulness, deployment and behavioural preferences, configurability and most important functionality of mobile crisis apps. Our results indicate that in addition to emergency and weather warnings, crime- and health-related warnings are also desired by many, as is the possibility for bidirectional communication. People also want one central app and are resistant to installing more than one crisis app. Furthermore, there are few significant differences between socioeconomic groups.

      May 2020  17th International Conference on Information Systems for Crisis Response and Management (ISCRAM 2020) Conference

    LoRa-based Device-to-Device Smartphone Communication for Crisis Scenarios

    Jonas Höchst, Lars Baumgärtner, Franz Kuntke, Alvar Penning, Artur Sterz, Bernd Freisleben

    PDF BibTeX

    Abstract
    In this paper, we present an approach to facilitate long-range device-to-device communication via smartphones in crisis scenarios. Through a custom firmware for low-cost LoRa capable micro-controller boards, called rf95modem, common devices for end users can be enabled to use LoRa through a Bluetooth, Wi-Fi, or serial connection. We present two applications utilizing the flexibility provided by the proposed firmware. First, we introduce a novel device-to-device LoRa chat application that works a) on the two major mobile platforms Android and iOS and b) on traditional computers like notebooks using a console-based interface. Second, we demonstrate how other infrastructure-less technology can benefit from our approach by integrating it into the DTN7 delay-tolerant networking software. The firmware, the device-to-device chat application, the integration into DTN7, as well as the experimental evaluation code fragments are available under permissive open-source licenses.

      April 2020  Proceedings of the International Conference on Wirtschaftsinformatik (WI) Conference

    Sticking with Landlines? Citizens’ Use and Perception of Social Media in Emergencies and Expectations Towards Emergency Services in Germany

    Jasmin Haunschild, Marc-André Kaufhold, Christian Reuter

    BibTeX DOI: 10.30844/wi_2020_o2-haunschild

    Abstract
    Crisis informatics has examined the use, potentials and weaknesses of social media in emergencies across different events (e.g., man-made, natural or hybrid), countries and heterogeneous participants (e.g., citizens or emergency services) for almost two decades. While most research analyzes specific cases, few studies have focused on citizens’ perceptions of different social media platforms in emergencies using a representative sample. Basing our questionnaire on a workshop with police officers, we present the results of a representative study on citizens’ perception of social media in emergencies that we conducted in Germany. Our study suggests that when it comes to emergencies, socio-demographic differences are largely insignificant and no clear preferences for emergency services’ social media strategies exist. Due to the widespread searching behavior on some platforms, emergency services can reach a wide audience by turning to certain channels but should account for groups with distinct preferences.

      April 2020  2020 CHI Conference on Human Factors in Computing Systems Conference

    Walk The Line: Leveraging Lateral Shifts of the Walking Path as an Input Modality for Head-Mounted Displays

    Florian Müller, Martin Schmitz, Daniel Schmitt, Sebastian Günther, Markus Funk, Max Mühlhäuser

    BibTeX DOI: 10.1145/3313831.3376852

    Abstract
    Recent technological advances have made head-mounted displays (HMDs) smaller and untethered, fostering the vision of ubiquitous interaction in a digitally augmented physical world. Consequently, a major part of the interaction with such devices will happen on the go, calling for interaction techniques that allow users to interact while walking. In this paper, we explore lateral shifts of the walking path as a hands-free input modality. The available input options are visualized as lanes on the ground parallel to the user’s walking path. Users can select options by shifting the walking path sideways to the respective lane. We contribute the results of a controlled experiment with 18 participants, confirming the viability of our approach for fast, accurate, and joyful interactions. Further, based on the findings of the controlled experiment, we present three example applications.

      March 2020  Technische Universität Darmstadt Darmstadt Thesis

    Secure device-to-device communication for emergency response

    Flor Álvarez

    PDF BibTeX DOI: 10.25534/tuprints-00011486

    Abstract
    Mobile devices have the potential to make a significant impact during disasters. However, their practical impact is severely limited by the loss of access to mobile communication infrastructure: Precisely, when there is a surge in demand for communications from people in a disaster zone, this capacity for communications is severely curtailed. This loss of communications undermines the effectiveness of the many recent innovations in the use of smartphones and similar devices to mitigate the effects of disasters. While various solutions have been proposed, e. g., by having handsets form wireless ad hoc networks, none are complete: Some are specific to certain mobile operating systems or operating system versions. Others result in unacceptably increased energy consumption, flattening the batteries of phones at a time when users need to conserve energy due to the loss of access to opportunities to recharge their mobile devices. Realistic user behaviour, including patterns of movement and communications, are also rarely addressed. Further, security is rarely considered in a comprehensive and satisfying manner, leaving users exposed to a variety of potential attacks. Thus there is a compelling need to find more effective solutions for communications, energy management, and security of mobile devices operating in disaster conditions. To address these shortcomings, this thesis provides a suite of comprehensive solutions that contribute to facilitate secure device-to-device communication for emergency response. This thesis works to solve these problems by: (i) Conducting a large-scale field-trial to understand and analyze civilians’ behaviour during disaster scenarios; (ii) Proposing a practical, lightweight scheme for bootstrapping device-to-device security, that is tailored for local urban operations representative of disaster scenarios; (iii) Realizing novel energy management strategies for the neighbour discovery problem, which deliver significant energy savings in return for only a minimal reduction in neighbour discovery efficiency; (iv) The description of novel concepts for using devices in a smart city environment that remain functional following a disaster to support communications among mobile devices. In short, this thesis adds considerably to the understanding of the difficulties in the formation of direct device-to-device communications networks composed primarily of civilians’ mobile devices, and how several facets of this problem can be mitigated. Several of the proposed enhancements are also implemented. Thus, this thesis also takes essential steps in the direction of realizing such solutions to demonstrate their feasibility on real devices, intending to improve the tools available to civilians post-disaster.

      February 2020  The Art, Science, and Engineering of Programming Article

    Implementing a Language for Distributed Systems: Choices and Experiences with Type Level and Macro Programming in Scala

    Pascal Weisenburger, Guido Salvaneschi

    PDF BibTeX DOI: 10.22152/programming-journal.org/2020/4/17

    Abstract
    Multitier programming languages reduce the complexity of developing distributed systems by developing the distributed system in a single coherent code base. The compiler or the runtime separate the code for the components of the distributed system, enabling abstraction over low level implementation details such as data representation, serialization and network protocols. Our ScalaLoci language allows developers to declare the different components and their architectural relation at the type level, allowing static reasoning about about distribution and remote communication and guaranteeing static type safety across components. The compiler splits the multitier program into the component-specific code and automatically generates the communication boilerplate. Communication between components can be modeled by declaratively specifying data flows between components using reactive programming. In this paper, we report on the implementation of our design and our experience with embedding our language features into Scala as a host language. We show how a combination of Scala’s advanced type level programming and its macro system can be used to enrich the language with new abstractions. We comment on the challenges we encountered and the solutions we developed for our current implementation and outline suggestions for an improved macro system to support the such use cases of embedding of domain-specific abstractions.

      February 2020  Embedded Wireless Systems and Networks (EWSN) Conference

    Improving the Reliability of Bluetooth Low Energy Connections

    Michael Spörk, Jiska Classen, Carlo Alberto Boano, Matthias Hollick, Kay Römer

    PDF BibTeX

    Abstract
    o sustain a reliable data exchange, applications based on Bluetooth Low Energy (BLE) need to effectively blacklist channels and adapt the physical mode of an active connection at runtime. Although the BLE specification foresees the use of these two mechanisms, their implementation is left up to the radio vendors and has not been studied in detail yet. This paper fills this gap: we first investigate experimentally how to assess the quality of a BLE connection at runtime using information gathered from the radio. We then show how this information can be used to promptly blacklist poor channels and select a physical mode that sustains a high link-layer reliability while minimizing power consumption. We implement both mechanisms on two popular platforms and show experimentally that they allow to significantly improve the reliability of BLE connections, with a reduction in packet loss by up to 22 % compared to existing solutions.

      February 2020  Technische Universität Darmstadt Darmstadt Thesis

    Availability by Design: Practical Denial-of-Service-Resilient Distributed Wireless Networks

    Milan Stute

    PDF BibTeX DOI: 10.25534/tuprints-00011457

    Abstract
    Distributed wireless networks (DWNs) where devices communicate directly without relying on Internet infrastructure are on the rise, driving new applications and paradigms such as multimedia, authentication, payment, Internet of things (IoT), vehicular communication, and emergency response. However, the increased societal reliance on technology and the resulting “always-on” expectations of the users increase the risk of denial-of-service (DoS) attacks as they can leverage disruption in new ways beyond extortions (ransomware) that are common in today’s Internet ecosystem. These new risks extend to our physical world, directly impacting our daily lives, e.g., by being locked out of a smart home or by disrupting vehicular collision avoidance systems. As a research community, we need to protect those new applications that—as we find—can be mapped to a total of three distinct networking scopes: neighbor, island, and archipelago. In this thesis, we advance the field in each of these scopes. First, we analyze two proprietary neighbor communication protocols, Apple Wireless Direct Link (AWDL) and Apple AirDrop, that are deployed on more than 1.4 billion devices worldwide. During the process, we uncover several security and privacy vulnerabilities ranging from design flaws to implementation bugs leading to a machine-in-the-middle (MitM) attack on AirDrop, a DoS attack on AWDL preventing communication, and DoS attacks enabling crashing of neighboring devices. In addition, we found privacy leaks that enable user identification and long-term tracking. All attacks can be mounted using low-cost off-the-shelf hardware. In total, we disclose eight distinct vulnerabilities that we mitigate in collaboration with Apple. Second, we design and implement a new island communication protocol tailored to IoT scenarios, which provides provable protections against previously neglected risks such as wormhole- and replay-supported greyhole attacks. We support our analytical findings with testbed experiments. Third, we propose an archipelago-scope communication framework for emergencies that achieves resiliency against flooding and Sybil attacks. We evaluate our design using an original expert knowledge-based simulation that models human mobility during the aftermath of the 2013 Typhoon Haiyan in the Philippines. Finally, and to nourish future research, we provide a guide for analyzing Apple’s wireless ecosystem and publish several software artifacts, including an AWDL Wireshark dissector, open AWDL and AirDrop implementations, a prototype of our IoT communication protocol, and our natural disaster mobility model.

      2020  BBK Bevölkerungsschutz Article

    Risikokulturen bei der Nutzung Sozialer Medien in Katastrophenlagen

    Christian Reuter, Marc-André Kaufhold, Stefka Schmid

    BibTeX

    Abstract
    Soziale Medien werden auf der ganzen Welt genutzt. Vergleicht man die allgemeine Nutzung sozialer Medien im Vereinigten Königreich (GB), Deutschland (DE), den Niederlanden (NL) und Italien (IT), zeigt sich, dass sie in Italien am wenigsten in Anspruch genommen werden. Dort sind knapp 40 % der Bevölkerung, d. h. 37 Millionen Menschen, in den sozialen Medien aktiv. Sowohl im Vereinigtem Königreich (59 %, 39 Mio.), Deutschland (55 %, 45 Mio.) als auch den Niederlanden (57 %, 9,74 Mio.) lassen sich ähnliche Tendenzen erkennen. Angesichts der Tatsache, dass mobile Endgeräte, die tendenziell immer griffbereit sind, sehr häufig zur Kommunikation über soziale Medien eingesetzt werden, ist es nicht verwunderlich, dass diese auch in Notsituationen genutzt werden [8]. Bis dato fehlt es an aussagekräftigen quantitativen und vergleichbaren Ergebnissen aus unterschiedlichen Ländern über die Wahrnehmung der Bevölkerung zur Nutzung von sozialen Medien in Notsituationen. Die im Folgenden vorgestellte Studie „The Impact of Risk Cultures: Citizens’ Perception of Social Media Use in Emergencies across Europe” [9] mit Beteiligung der TU Darmstadt, Universität Siegen und dem Tavistock Institute (London), möchte das bestehende Defizit adressieren. Anhand der repräsentativen Umfrageergebnisse werden zunächst vier europäische Länder präsentiert und dann miteinander verglichen. Ziel ist es, Ähnlichkeiten und Unterschiede in der Nutzung sozialer Medien in Notsituationen zu erfassen. Frühere Forschungsergebnisse haben im Hinblick auf Katastrophen gezeigt, dass es unterschiedliche Risikokulturen in europäischen Ländern gibt, die das Verhalten der Bevölkerung jeweils unterschiedlich beeinflussen und prägen (vgl. [4]; [5]).

      2020  Disaster Research Days 2020 Conference

    Politische Naturkatastrophen - Wie Ideologien den Umgang mit Hochwasser prägten

    Nadja Thiessen

    BibTeX

    Abstract
    Die Bewältigung von Krisen und Katastrophen weist stets auch eine politische Dimension auf. Exemplarisch lässt sich dies für das 20. Jahrhundert in Deutschland und den dortigen Umgang mit Flusshochwassern nachweisen. In den ausgewählten Fallstädten Mannheim und Dresden konnte zwar ein konstanter Bewältigungskreislauf im städtischen Hochwasserschutz identifiziert werden, wenn dieser jedoch in seinen historischen Kontext gesetzt wird, offenbaren sich Unterschiede. Sie können vor allem auf die verschiedenen politischen Rahmenbedingungen wie Staatsform und vorherrschende Ideologie zurückgeführt werden. Im folgenden Beitrag werden die Einflüsse historischer autoritärer Regime, insbesondere des Nationalsozialismus und der DDR, den demokratischen Systemen gegenübergestellt.

      2020  2020 CHI Conference on Human Factors in Computing Systems (CHI ‘20) Conference

    Podoportation: Foot-Based Locomotion in Virtual Reality

    Julius von Willich, Martin Schmitz, Florian Müller, Daniel Schmitt, Max Mühlhäuser

    BibTeX DOI: 10.1145/3313831.3376626

    Abstract
    Virtual Reality (VR) allows for infinitely large environments. However, the physical traversable space is always limited by real-world boundaries. This discrepancy between physical and virtual dimensions renders traditional locomotion methods used in real world unfeasible. To alleviate these limitations, research proposed various artificial locomotion concepts such as teleportation, treadmills, and redirected walking. However, these concepts occupy the user’s hands, require complex hardware or large physical spaces. In this paper, we contribute nine VR locomotion concepts for foot-based and hands-free locomotion, relying on the 3D position of the user’s feet and the pressure applied to the sole as input modalities. We evaluate our concepts and compare them to state-of-the-art point & teleport technique in a controlled experiment with 20 participants. The results confirm the viability of our approaches for hands-free and engaging locomotion. Further, based on the findings, we contribute a wireless hardware prototype implementation.

      January 2020  Information Processing & Management Article

    Rapid relevance classification of social media posts in disasters and emergencies: A system and evaluation featuring active, incremental and online learning

    Marc-André Kaufhold, Markus Bayer, Christian Reuter

    BibTeX DOI: 10.1016/j.ipm.2019.102132

    Abstract
    The research field of crisis informatics examines, amongst others, the potentials and barriers of social media use during disasters and emergencies. Social media allow emergency services to receive valuable information (e.g., eyewitness reports, pictures, or videos) from social media. However, the vast amount of data generated during large-scale incidents can lead to issue of information overload. Research indicates that supervised machine learning techniques are sui- table for identifying relevant messages and filter out irrelevant messages, thus mitigating in- formation overload. Still, they require a considerable amount of labeled data, clear criteria for relevance classification, a usable interface to facilitate the labeling process and a mechanism to rapidly deploy retrained classifiers. To overcome these issues, we present (1) a system for social media monitoring, analysis and relevance classification, (2) abstract and precise criteria for re- levance classification in social media during disasters and emergencies, (3) the evaluation of a well-performing Random Forest algorithm for relevance classification incorporating metadata from social media into a batch learning approach (e.g., 91.28%/89.19% accuracy, 98.3%/89.6% precision and 80.4%/87.5% recall with a fast training time with feature subset selection on the European floods/BASF SE incident datasets), as well as (4) an approach and preliminary eva- luation for relevance classification including active, incremental and online learning to reduce the amount of required labeled data and to correct misclassifications of the algorithm by feed- back classification. Using the latter approach, we achieved a well-performing classifier based on the European floods dataset by only requiring a quarter of labeled data compared to the tradi- tional batch learning approach. Despite a lesser effect on the BASF SE incident dataset, still a substantial improvement could be determined.

      2020  International Journal of Disaster Risk Reduction Article

    Empirical insights for designing Information and Communication Technology for International Disaster Response

    Milan Stute, Max Maass, Tom Schons, Marc-André Kaufhold, Christian Reuter, Matthias Hollick

    BibTeX DOI: 10.1016/j.ijdrr.2020.101598

    Abstract
    Due to the increase in natural disasters in the past years, Disaster Response Organizations (DROs) are faced with the challenge of coping with more and larger operations. Currently appointed Information and Communications Technology (ICT) used for coordination and communication is sometimes outdated and does not scale, while novel technologies have the potential to greatly improve disaster response efficiency. To allow adoption of these novel technologies, ICT system designers have to take into account the particular needs of DROs and characteristics of International Disaster Response (IDR). This work attempts to bring the humanitarian and ICT communities closer together. In this work, we analyze IDR-related documents and conduct expert interviews. Using open coding, we extract empirical insights and translate the peculiarities of DRO coordination and operation into tangible ICT design requirements. This information is based on interviews with active IDR staff as well as DRO guidelines and reports. Ultimately, the goal of this paper is to serve as a reference for future ICT research endeavors to support and increase the efficiency of IDR operations.

      November 2019  ASE2019 Conference

    Automated Refactoring to Reactive Programming

    Mirco Köhler, Guido Salvaneschi

    PDF BibTeX DOI: 10.1109/ASE.2019.00082

    Abstract
    Reactive programming languages and libraries, such as ReactiveX, have been shown to significantly improve software design and have seen important industrial adoption over the last years. Asynchronous applications – which are notoriously error-prone to implement and to maintain – greatly benefit from reactive programming because they can be defined in a declarative style, which improves code clarity and extensibility. In this paper, we tackle the problem of refactoring existing code bases that are designed using traditional abstractions for asynchronous programming. We propose 2Rx, a refactoring tool to automatically convert asynchronous code to reactive programming. Our evaluation on top-starred GitHub projects shows that 2Rx is effective with the most common asynchronous constructs, covering ~94.7% of the projects with asynchronous computations, and it can provide a refactoring for ~91.7% of their occurrences.

      October 2019  13th International Workshop on Wireless Network Testbeds, Experimental Evaluation & Characterization (WiNTECH ’19) Conference

    Free Your CSI: A Channel State Information Extraction Platform For Modern Wi-Fi Chipsets

    Francesco Gringoli, Matthias Schulz, Jakob Link, Matthias Hollick

    PDF BibTeX DOI: 10.1145/3349623.3355477

    Abstract
    Modern wireless transmission systems heavily benefit from knowing the channel response. The evaluation of Channel State Information (CSI) during the reception of a frame preamble is fundamental to properly equalizing the rest of the transmission at the receiver side. Reporting this state information back to the transmitter facilitates mechanisms such as beamforming and MIMO, thus boosting the network performance. While these features are an integral part of standards such as 802.11ac, accessing CSI data on commercial devices is either not possible, limited to outdated chipsets or very inflexible. This hinders the research and development of innovative CSI-dependent techniques including localization, object tracking, and interference evaluation. To help researchers and practitioners, we introduce the nexmon CSI Extractor Tool. It allows per-frame CSI extraction for up to four spatial streams using up to four receive chains on modern Broadcom and Cypress Wi-Fi chips with up to 80MHz bandwidth in both the 2.4 and 5GHz bands. The tool supports devices ranging from the low-cost Raspberry Pi platform, over mobile platforms such as Nexus smartphones to state-of-the-art Wi-Fi APs. We release all tools and Wi-Fi firmware patches as extensible open source project. It includes our user-friendly smartphone application to demonstrate the CSI extraction capabilities in form of a waterfall diagram.

      October 2019  2019 IEEE Global Humanitarian Technology Conference (GHTC 2019) Conference

    Smart Street Lights and Mobile Citizen Apps for Resilient Communication in a Digital City

    Lars Baumgärtner, Jonas Höchst, Patrick Lampe, Ragnar Mogk, Artur Sterz, Pascal Weisenburger, Mira Mezini, Bernd Freisleben

    BibTeX

    Abstract
    Currently, nearly four billion people live in urban areas. Since this trend is increasing, natural disasters or terrorist attacks in such areas affect an increasing number of people. While information and communication technology is crucial for the operation of urban infrastructures and the well-being of its inhabitants, current technology is quite vulnerable to disruptions of various kinds. In future smart cities, a more resilient urban infrastructure is imperative to handle the increasing number of hazardous situations. We present a novel resilient communication approach based on smart street lights as part of the public infrastructure. It supports people in their everyday life and adapts its functionality to the challenges of emergency situations. Our approach relies on various environmental sensors and in-situ processing for automatic situation assessment, and a range of communication mechanisms (e.g., public WiFi hotspot functionality and mesh networking) for maintaining a communication network. Furthermore, resilience is not only achieved based on infrastructure deployed by a digital city’s municipality, but also based on integrating citizens through software that runs on their mobile devices (e.g., smartphones and tablets). Web-based zero-installation and platform-agnostic apps can switch to device-to-device communication to continue benefiting people even during a disaster situation. Our approach, featuring a covert channel for professional responders and the zero-installation app, is evaluated through a prototype implementation based on a commercially available street light.

      October 2019  44th IEEE Conference on Local Computer Networks (LCN 2019) Conference

    OPPLOAD: Offloading Computational Workflows in Opportunistic Networks

    Artur Sterz, Lars Baumgärtner, Jonas Höchst, Patrick Lampe, Bernd Freisleben

    BibTeX

    Abstract
    Computation offloading is often used in mobile cloud computing, edge computing, and/or fog computing to cope with resource limitations of mobile devices in terms of computational power, storage, and energy. Computation offloading is particularly challenging in situations where network connectivity is periodic, intermittent, or error-prone. In this paper, we present OPPLOAD, a novel framework designed for offloading computational workflows in opportunistic networks that provide support for communication in such situations. The individual tasks forming a workflow can be assigned to particular remote execution platforms, called workers, either preselected ahead of time or decided just in time where a matching worker will automatically be assigned for the next task in the workflow. Workers announce their capabilities, i.e., tasks are only assigned to capable workers. Furthermore, tasks of a workflow can be executed on multiple workers that are automatically selected to balance the overall load. OPPLOAD also offers the ability to handle several types of error and exceptions appropriately. Our Python implementation of OPPLOAD, which uses the Serval Mesh to handle networking and routing, is publicly available as open source software. The results of our experimental evaluation demonstrate the feasibility of our approach.

      October 2019  18th International Conference on Ad Hoc Networks and Wireless (ADHOC-NOW 2019) Conference

    DTN7: An Open-Source Disruption-tolerant Networking Implementation of Bundle Protocol 7

    Alvar Penning, Lars Baumgärtner, Jonas Höchst, Artur Sterz, Mira Mezini, Bernd Freisleben

    PDF BibTeX

    Abstract
    In disruption-tolerant networking (DTN), data is transmitted in a store-carry-forward fashion from network node to network node. In this paper, we present an open source DTN implementation, called DTN7, of the recently released Bundle Protocol Version 7 (draft version 13). DTN7 is written in Go and provides features like memory safety and concurrent execution. With its modular design and interchangeable components, DTN7 facilitates DTN research and application development. Furthermore, we present results of a comparative experimental evaluation of DTN7 and other DTN systems including Serval, IBR-DTN, and Forban. Our results indicate that DTN7 is a flexible and efficient open-source multi-platform implementation of the most recent Bundle Protocol Version 7.

      October 2019  44th IEEE Conference on Local Computer Networks (LCN) Conference

    Multi-Strategy Simulation of Aerial Post-Disaster Ad Hoc Communication Support Systems

    Julian Zobel, Patrick Lieser, Ralf Steinmetz

    PDF BibTeX

    Abstract
    In case of destroyed or impaired infrastructure due to natural catastrophes, mobile devices such as smartphones can be used to create civilian ad hoc networks to provide basic means of communication. Due to the human behavior to form groups and cluster around significant locations in such situations, however, the network is often heavily intermittent, and thus, communication between clusters is impossible. Aerial Post-Disaster Ad Hoc Communication Support Systems can overcome the gaps between clusters, but the performance is highly dependent on factors like the applied strategy, the amount of UAVs, or their technical specifications. In this demonstration, we present different support strategies in an urban post-disaster scenario. Attendees can interact and select strategies and explore different strategy parameter settings, while observing the effect on the network performance and, additionally, gaining a comprehensive insight into the strategy behavior. The interaction with the demonstration underlines the vast amount of different settings and influence factors, an aerial system operator must take into account when selecting and adapting a strategy suitable for the current situation, as motivated in our accompanying main conference pape ZLD+19.

      August 2019  2019 World Congress on Resilience, Reliability and Asset Management (WCRRAM) Conference

    The Emergency Responsive Digital City

    Matthias Hollick, Anne Hofmeister, Jens Ivo Engels, Bernd Freisleben, Gerrit Hornung, Anja Klein, Michèle Knodt, Imke Lorenz, Max Mühlhäuser, Peter F. Pelz, Annette Rudolph-Cleff, Ralf Steinmetz, Florian Steinke, Oskar von Stryk

    BibTeX

      2019  Proceedings of the ACM on Programming Languages Article

    A Fault-Tolerant Programming Model for Distributed Interactive Applications

    Ragnar Mogk, Joscha Drechsler, Guido Salvaneschi, Mira Mezini

    BibTeX DOI: 10.1145/3360570

    Abstract
    Ubiquitous connectivity of web, mobile, and IoT computing platforms has fostered a variety of distributed applications with decentralized state. These applications execute across multiple devices with varying reliability and connectivity. Unfortunately, there is no declarative fault-tolerant programming model for distributed interactive applications with an inherently decentralized system model. We present a novel approach to automating fault tolerance using high-level programming abstractions tailored to the needs of distributed interactive applications. Specifically, we propose a calculus that enables formal reasoning about applications’ dataflow within and across individual devices. Our calculus reinterprets the functional reactive programming model to seamlessly integrate its automated state change propagation with automated crash recovery of device-local dataflow and disconnection-tolerant distribution with guaranteed automated eventual consistency semantics based on conflict-free replicated datatypes. As a result, programmers are relieved of handling intricate details of distributing change propagation and coping with distribution failures in the presence of interactivity. We also provides proofs of our claims, an implementation of our calculus, and an empirical evaluation using a common interactive application.

      2019  Proceedings of the ACM on Programming Languages Article

    Language-Integrated Privacy-Aware Distributed Queries

    Guido Salvaneschi, Mirko Köhler, Daniel Sokolowski, Philipp Haller, Sebastian Erdweg, Mira Mezini

    BibTeX DOI: 10.1145/3360593

    Abstract
    Distributed query processing is an effective means for processing large amounts of data. To abstract from the technicalities of distributed systems, algorithms for operator placement automatically distribute sequential data queries over the available processing units. However, current algorithms for operator placement focus on performance and ignore privacy concerns that arise when handling sensitive data. We present a new methodology for privacy-aware operator placement that both prevents leakage of sensitive information and improves performance. Crucially, our approach is based on an information-flow type system for data queries to reason about the sensitivity of query subcomputations. Our solution unfolds in two phases. First, placement space reduction generates deployment candidates based on privacy constraints using a syntax-directed transformation driven by the information-flow type system. Second, constraint solving selects the best placement among the candidates based on a cost model that maximizes performance. We verify that our algorithm preserves the sequential behavior of queries and prevents leakage of sensitive data. We implemented the type system and placement algorithm for a new query language SecQL and demonstrate significant performance improvements in benchmarks.

    Software and Tools

    DTN7-go

    In disruption-tolerant networking (DTN), data is transmitted in a store-carry-forward fashion from network node to network node. We are presenting free and open source DTN implementations of the recently released Bundle Protocol Version 7. DTN7-go is written in Go and provides features like memory safety and concurrent execution.

    Source Code Project Website

    DTN7-rs

    Rust implementation of a daemon for DTN7 Bundle Protocol draft.

    Source Code Buschfunk Project

    OpenDrop

    OpenDrop is a command-line tool written in Python that allows sharing files between devices directly over Wi-Fi. Its unique feature is that it is protocol-compatible with Apple AirDrop which allows to share files with Apple devices running iOS and macOS.

    Source Code Python Package Project Website

    OWL

    Open Wireless Link (OWL) is an open implementation of the Apple Wireless Direct Link (AWDL) ad hoc protocol for Linux and macOS written in C.

    Source Code Project Website

    OpenHaystack

    OpenHaystack is a framework for tracking personal Bluetooth devices via Apple’s massive Find My network. Use it to create your own tracking tags that you can append to physical objects (keyrings, backpacks, …) or integrate it into other Bluetooth-capable devices such as notebooks.

    Source Code Project Website

    ChirPOTLE

    ChirpOTLE is a practical LoRaWAN security evaluation framework that provides tools for the deployment and management of a LoRa testbed based on COTS hardware. It allows managing LoRa field nodes from a central controller and to orchestrate experiments and tests using a Python 3 interface.

    Source Code Paper

    AdapterHub

    “Adapter” refers to a set of newly introduced weights, typically within the layers of a transformer model. Adapters provide an alternative to fully fine-tuning the model for each downstream task, while maintaining performance.

    Project Website Demonstration

    SPARCODE

    We proposed SPARCODE, a community detection method that uses spectral partitioning based on estimating a robust and sparse graph model.

    Source Code

    image_projection

    image_projection is a ROS package to create various projections from multiple calibrated cameras.

    Source Code

    PrivateDrop

    We developed a solution named PrivateDrop to replace the flawed original AirDrop design. PrivateDrop is based on optimized cryptographic private set intersection protocols that can securely perform the contact discovery process between two users without exchanging vulnerable hash values.

    Source Code Project Website

    T-LARS

    Computes the solution path of the Terminating-LARS (T-LARS) algorithm, which is a major building block of the T-Rex Selector.

    Source Code Project Website

    TRexSelector

    Performs fast variable selection in large-scale high-dimensional settings while controlling the false discovery rate (FDR) at a user-defined target level.

    Source Code Project Website

    1-stage-wseg

    Single-stage semantic segmentation from image labels

    Source Code

    da-sac

    Self-supervised augmentation consistency for adapting semantic segmentation

    Source Code

    dense-ulern-vos

    Dense unsupervised learning for video segmentation

    Source Code

    coreemu-lab

    coreemu-lab (short: clab) is an environment for automated evaluation of networking software. It should work headless on any platform that can run Docker and with X11 it has been successfully tested on Linux and macOS.

    Source Code Paper

    REScala

    REScala is a set of libraries for principled reactive programming, a programming paradigm that integrates the strong guarantees of functional reactive programming into object oriented programs, and is useful in many contexts such as traditional user interfaces, concurrent and distributed applications, web development, and server software.

    Source Code Project Website User Manual

    rf95modem MCU firmware

    This project provides a modem firmware for microcontroller boards with a RF95 compatible radio module and a serial interface such as the adafruit feather m0 lora device or the heltec oled lora 32 modules. On various ESP32 based boards optional features such as OLED status display, GPS, BLE or WiFi support can be enabled.

    Source Code