TY - GEN A1 - Dietrich, Lucas A1 - Förster, Benjamin A1 - Langendörfer, Peter A1 - Hinze, Thomas T1 - A visual discrete event-based simulator for protection of plants against herbivores employed as computational optimization game N2 - Plants come with sophisticated strategies to survive within a highly competing environment. In addition, they need to resist frequent attacks from a variety of herbivores acting alone, in small groups, or in swarms. Since the amount of energy a plant might invest in defense and reproduction is limited, a complex optimization problem emerges. In a shared habitat, plants fight herbivores by shape and camouflage, by the release of specific toxins, or by attracting predators of herbivores. Furthermore, plants alert their surrounding field by signaling substances in the event of an assault. Transported by air or through a network of roots, signaling substances reach neighbors to trigger their defense. The offsprings of a plant commonly grow within a certain distance to benefit from symbiotic protection. We introduce a grid-based visual simulation software for detailed configuration and subsequent processing of the behavior of the resulting system in time and space. In terms of solution to a computational optimization problem inspired by nature, settings with low energy need and long life able to cope with different patterns of attack can be figured out and analyzed. Applications include novel techniques for efficient construction and secure operation of sensor networks. KW - Plant defense KW - Event-based simulation KW - Topological optimization Y1 - 2025 U6 - https://doi.org/10.48550/arXiv.2509.15787 SP - 1 EP - 25 PB - arXiv ER - TY - GEN A1 - Lehniger, Kai A1 - Langendörfer, Peter T1 - Investigating compact shadow stacks for the Xtensa LX architecture T2 - 2025 14th Mediterranean Conference on Embedded Computing (MECO) N2 - This paper presents a compact shadow stack implementation for the Xtensa LX architecture, implemented for an ESP32 microcontroller. The implementation is done by changing the register window exception handlers in the FreeRTOS operating system. In a second implementation the shadow stack was extended to all register values, not just return addresses. This not only increases security but also performance, reducing the overhead from 17.220% to 13.529%, in the worst-case-scenario. The paper also suggests future improvements that could decrease the overhead even further. KW - Xtensa KW - ESP32 KW - Return address protection KW - Shadow stack Y1 - 2025 SN - 979-8-3315-1341-2 U6 - https://doi.org/10.1109/MECO66322.2025.11049161 SP - 1 EP - 6 PB - IEEE CY - Piscataway, NJ ER - TY - GEN A1 - Alsabbagh, Wael A1 - Sayegh, Bahij A1 - Kim, Chaerin A1 - Langendorfer, Peter T1 - FeatNet-IDS : anomaly detection based-features for industrial internet of things systems T2 - 2025 Silicon Valley Cybersecurity Conference (SVCC) N2 - Anomaly detection is critical for securing Industrial Internet of Things (IIoT) systems against evolving cyber threats. This paper introduces FeatNet-IDS, an enhanced Intrusion Detection System (IDS) for MQTT-based IIoT environments. FeatNet- IDS combines advanced feature selection and machine learning techniques to improve anomaly detection accuracy, efficiency, and interpretability. Evaluated on a domain-specific dataset of MQTT-related attacks, including control logic injection, the framework achieves 92.3% accuracy, 91.5% F1-score, and 96.8% ROC AUC, with a low false alarm rate of 2.73%. By emphasizing feature stability and importance, FeatNet-IDS outperforms or matches state-of-the-art methods, demonstrating robustness and transparency. Practical deployment implications are discussed, with future work addressing dataset expansion, hyperparameter tuning, and advanced ensemble approaches. The dataset and framework codes are publicly available to support further research. KW - IIoT KW - MQTT Protocol KW - Cyberattacks KW - Cybersecurity KW - Anomaly Detection Y1 - 2025 SN - 979-8-3315-3429-5 U6 - https://doi.org/10.1109/SVCC65277.2025.11133650 SP - 1 EP - 8 PB - IEEE CY - Piscataway, NJ ER - TY - GEN A1 - Alsabbagh, Wael A1 - Urrego, Diego A1 - Langendörfer, Peter T1 - Smart traps for smart systems : scalable honeynets for IIoT cybersecurity T2 - 2025 34th International Conference on Computer Communications and Networks (ICCCN) N2 - Honeypots serve as decoy systems that attract and monitor intruders, offering insights into their behavior. When interconnected, these honeypots form honeynets, simulating high-value environments to engage attackers and facilitate deeper analysis. However, in Industrial Internet of Things (IIoT) networks, deploying honeypots presents challenges such as static configurations, lack of network context, and difficulties in scaling.In this paper, we introduce TrapNet, a scalable, lightweight honeynet framework specifically designed for IIoT environments. TrapNet combines compact, on-site honeypots with large-scale, adaptive honeynets deployed on fog and cloud infrastructures using microservices. This approach enables fast deployment, scalability, and flexibility, providing an effective solution for IIoT cybersecurity. Our experimental results show that TrapNet efficiently detects intrusions while maintaining low resource overhead. Additionally, all code and configurations used in this study are publicly available, fostering further research and innovation in honeynet design and IIoT security. KW - Honeypots KW - Cybersecurity KW - IIoT Security KW - Fog Computing KW - Cloud Infrastructure Y1 - 2025 SN - 979-8-3315-0898-2 U6 - https://doi.org/10.1109/ICCCN65249.2025.11133761 SP - 1 EP - 6 PB - IEEE CY - Piscataway, NJ ER - TY - GEN A1 - Yener, Onur A1 - Brzozowski, Marcin A1 - Chitauro, Ronald A1 - Langendörfer, Peter T1 - Pushing the limits of 5G private networks : a practical examination of network stressors T2 - 2025 International Conference on Computing, Networking and Communications (ICNC) N2 - The adoption of 5G private networks is revolutionizing industrial connectivity, yet it brings numerous challenges that need addressing to ensure safety and reliability. Our research utilizes an OpenAirInterface (OAI) based 5G testbed to explore and mitigate these challenges. We particularly focus on stressors such as jamming to evaluate network resilience. Our methodology involves setting up a comprehensive testbed, collecting data under various stress conditions, and implementing solutions. Initial findings demonstrate that small environmental changes and structural interferences significantly impact network performance. We observed that indoor variables and uplink communication pose substantial challenges, underscoring the need for advanced algorithms to detect and mitigate jamming attacks in dynamic scenarios. Our results highlight the importance of real-time spectrum analysis and adaptive network management in maintaining optimal performance. This study aims to contribute to the development of robust, resilient 5G networks suitable for complex industrial environments. KW - 5G Private Networks KW - OpenAirInterface KW - Jamming KW - Network Stressors KW - Industrial Connectivity KW - Network Resilience Y1 - 2025 SN - 979-8-3315-2096-0 U6 - https://doi.org/10.1109/ICNC64010.2025.10993913 SP - 752 EP - 756 PB - IEEE CY - Piscataway, NJ ER - TY - GEN A1 - Pidvalnyi, Illia A1 - Kostenko, Anna A1 - Sudakov, Oleksandr A1 - Isaev, Dmytro A1 - Maximyuk, Oleksandr A1 - Krishtal, Oleg A1 - Iegorova, Olena A1 - Kabin, Ievgen A1 - Dyka, Zoya A1 - Ortmann, Steffen A1 - Langendörfer, Peter T1 - Classification of epileptic seizures by simple machine learning techniques : application to animals’ electroencephalography signals T2 - IEEE access N2 - Detection and prediction of the onset of seizures are among the most challenging problems in epilepsy diagnostics and treatment. Small electronic devices capable of doing that will improve the quality of life for epilepsy patients while also open new opportunities for pharmacological intervention. This paper presents a novel approach using machine learning techniques to detect seizures onset using intracranial electroencephalography (EEG) signals. The proposed approach was tested on intracranial EEG data recorded in rats with pilocarpine model of temporal lobe epilepsy. A principal component analysis was applied for feature selection before using a support vector machine for the detection of seizures. Hjorth’s parameters and Daubechies discrete wavelet transform coefficients were found to be the most informative features of EEG data. We found that the support vector machine approach had a classification sensitivity of 90% and a specificity of 74% for detecting ictal episodes. Changing the epoch parameter from one to twenty-one seconds results in changing the redistribution of principal components’ values to 10% but does not affect the classification result. Support vector machines are accessible and convenient methods for classification that have achieved promising classification quality, and are rather lightweight compared to other machine learning methods. So we suggest their future use in mobile devices for early epileptic seizure and preictal episode detection. KW - Epilepsy KW - Single-channel intracranial encephalographic data KW - PCA KW - SVM KW - Automated system KW - Rats Y1 - 2025 U6 - https://doi.org/10.1109/ACCESS.2025.3527866 SN - 2169-3536 VL - 13 SP - 8951 EP - 8962 PB - Institute of Electrical and Electronics Engineers (IEEE) CY - Piscataway, NJ ER - TY - GEN A1 - Aftowicz, Marcin A1 - Fritscher, Markus A1 - Lehniger, Kai A1 - Wenger, Christian A1 - Langendörfer, Peter A1 - Brzozowski, Marcin T1 - Hardware-friendly Nyström approximation for water treatment anomaly detection T2 - IECON 2024 - 50th Annual Conference of the IEEE Industrial Electronics Society : proceedings N2 - This paper presents an approach to accelerate One-Class Support Vector Machines (SVM) using a hardware-friendly kernel that doesn't rely on multiplication operations, thus adaptable to hardware platforms. Leveraging Nyström approximation, we implemented a pipeline and compared its performance against a software implementation using libsvm. Furthermore, we evaluated the efficiency of our approach by deploying it on an FPGA. Our experiments, conducted on the SWaT dataset, demonstrate a 50x speedup using the FPGA implementation, achieving a classification time of 21 microseconds per instance. Importantly, we find no degradation in performance, as measured by the f-score of the attack class in the test set. This study explores the potential of hardware acceleration in optimizing anomaly detection systems for real-time applications. KW - OCSVM KW - FPGA KW - Nyström KW - Anomaly Detection KW - SWaT Y1 - 2024 SN - 978-1-6654-6454-3 U6 - https://doi.org/10.1109/IECON55916.2024.10905880 SP - 1 EP - 7 PB - Institute of Electrical and Electronics Engineers (IEEE) CY - Piscataway, NJ ER - TY - GEN A1 - Lehniger, Kai A1 - Langendorfer, Peter T1 - Elevating parallel shadow stack concepts for architectures with register windows T2 - IEEE transactions on dependable and secure computing N2 - Register windows are a processor feature that was originally developed to reduce the frequency for register spilling. This paper investigates how this mechanism can be used to implement a parallel shadow stack to protect return addresses from being corrupted. Furthermore, multiple improvements based on properties directly following the usage of register windows are proposed. Different configurations of parallel shadow stacks are evaluated based on prototype implementations on an ESP32 microcontroller and tested with different benchmarks. Performance overheads below 0.01% for CoreMark, below 3% for the worst case in MbedsTLS, and a worst case scenario of 22.153% for a recursive function show that our concepts are able to be efficiently developed even for embedded devices like the ESP32. KW - Buffer overflow KW - Shadow stack KW - Return-oriented programming KW - Register windows KW - Xtensa KW - ESP32 Y1 - 2025 U6 - https://doi.org/10.1109/TDSC.2025.3604556 SN - 1545-5971 SN - 1941-0018 SP - 1 EP - 18 PB - Institute of Electrical and Electronics Engineers (IEEE) CY - Piscataway, NJ ER -