Enhancing IoT System Reliability through Integrated Anomaly Detection and Sensor Trust Modeling Using Deep Learning
AUTHORS
Youk Greeve,PhD. Student, University of Gothenburg, Sweden
Cruise Speck,PhD. Student, University of Gothenburg, Sweden
ABSTRACT
The rapid deployment of Internet of Things (IoT) infrastructures in advanced digital economies such as Sweden has introduced critical engineering challenges related to data integrity, system reliability, and cybersecurity. As IoT systems increasingly support essential applications—including smart grids, intelligent transportation systems, and sustainable urban infrastructure—the trustworthiness of sensor data becomes a key requirement for safe and efficient operation. However, conventional intrusion detection approaches primarily focus on network-level threats and often overlook validating sensor data, particularly in resource-constrained and heterogeneous IoT environments. This study proposes a Behavior Detection Mechanism for Trust Sensor Data (BDM-TSD), a deep learning-based framework designed to simultaneously detect malicious network behavior and assess the reliability of multi-sensor data streams. The framework integrates packet-level analysis with sensor behavior modeling, leveraging long short-term memory (LSTM) networks to capture temporal dependencies in both communication traffic and sensing data. By incorporating features such as packet metadata, device operational states, and time-series sensing characteristics, the proposed approach enables the identification of anomalies caused by both external cyberattacks and internal data manipulation. Unlike conventional methods that rely on predefined attack signatures, the proposed framework emphasizes adaptive learning and behavioral profiling, making it well-suited to dynamic, large-scale IoT deployments. In addition, the model is designed for computational efficiency, enabling deployment on resource-constrained edge devices without significant performance overhead. Experimental results demonstrate that the proposed method achieves high detection accuracy while maintaining low false-positive and false-negative rates. The findings indicate that integrating sensor-level trust evaluation with network-level anomaly detection significantly enhances overall system resilience. This work contributes to the development of secure IoT engineering solutions by providing a scalable and practical approach to improving the reliability of sensor-driven systems in Sweden and comparable digitally advanced environments.
KEYWORDS
Internet of Things (IoT), Anomaly detection, Deep learning, Sensor data trust, LSTM networks, IoT security
REFERENCES
[1] D. E. Okonta and V. Vukovic, “Smart cities software applications for sustainability and resilience,” Heliyon, vol. 10, no. 12, e32654, (2024). DOI:10.1016/j.heliyon.2024.e32654(CrossRef)(Google Scholar)
[2] E. Boffa and A. Maffei, “Investigating the impact of digital transformation on manufacturers’ business model: Insights from Swedish industry,” Journal of Open Innovation: Technology, Market, and Complexity, vol. 10, no. 2, 100312, (2024). DOI:10.1016/j.joitmc.2024.100312(CrossRef)(Google Scholar)
[3] K. K. Patel and S. M. Patel, “Internet of Things-IoT: Definition, characteristics, architecture, enabling technologies, application and future challenges,” International Journal of Engineering Science and Computing, vol. 6, pp. 6122–6131, (2016)
[4] S. Sicari, A. Rizzardi, L. Grieco, and A. Coen-Porisini, “Security, privacy and trust in Internet of Things: The road ahead,” Computer Networks, vol. 76, pp. 146–164, (2015). DOI:10.1016/j.comnet.2014.11.008(CrossRef)(Google Scholar)
[5] E. Borgia, “The Internet of Things vision: Key features, applications and open issues,” Computer Communications, vol. 54, pp. 1–31, (2014). DOI:10.1016/j.comcom.2014.09.008(CrossRef)(Google Scholar)
[6] Y. Meidan, M. Bohadana, Y. Mathov, Y. Mirsky, A. Shabtai, D. Breitenbacher, and Y. Elovici, “N-BaIoT—Network-based detection of IoT botnet attacks using deep autoencoders,” IEEE Pervasive Computing, vol. 17, no. 3, pp. 12–22, (2018). DOI:10.1109/MPRV.2018.03367731(CrossRef)(Google Scholar)
[7] H. Hindy, D. Brosset, E. Bayne, A. K. Seeam, C. Tachtatzis, R. Atkinson, and X. Bellekens, “A taxonomy of network threats and the effect of current datasets on intrusion detection systems,” IEEE Access, vol. 8, pp. 104650–104675, (2020). DOI:10.1109/ACCESS.2020.3000179(CrossRef)(Google Scholar)
[8] S. Liang, S. Jin, and Y. Chen, “A review of edge computing technology and its applications in power systems,” Energies, vol. 17, no. 13, 3230, (2023). DOI:10.3390/en17133230(CrossRef)(Google Scholar)
[9] M. L. Ali, K. Thakur, S. Schmeelk, J. Debello, and D. Dragos, “Deep learning vs. machine learning for intrusion detection in computer networks: A comparative study,” Applied Sciences, vol. 15, no. 4, 1903, (2024). DOI:10.3390/app15041903(CrossRef)(Google Scholar)
[10] A. Abeshu and N. Chilamkurti, “Deep learning: The frontier for distributed attack detection in fog-to-things computing,” IEEE Communications Magazine, vol. 56, no. 2, pp. 169–175, (2018). DOI:10.1109/MCOM.2018.1700332(CrossRef)(Google Scholar)
[11] H. Bangui and B. Buhnova, “Lightweight intrusion detection for edge computing networks using deep forest and bio-inspired algorithms,” Computers and Electrical Engineering, vol. 100, 107901, (2022). DOI:10.1016/j.compeleceng.2022.107901(CrossRef)(Google Scholar)
[12] Z. Yan, P. Zhang, and A. V. Vasilakos, “A survey on trust management for Internet of Things,” Journal of Network and Computer Applications, vol. 42, pp. 120–134, (2014). DOI:10.1016/j.jnca.2014.01.014(CrossRef)(Google Scholar)
[13] U. C. Akuthota and L. Bhargava, “Transformer-based intrusion detection for IoT networks,” IEEE Internet of Things Journal, vol. 12, no. 5, pp. 6062–6067, (2025). DOI:10.1109/JIOT.2025.3525494(CrossRef)(Google Scholar)
[14] M. Gao, L. Wu, Q. Li, and W. Chen, “Anomaly traffic detection in IoT security using graph neural networks,” Journal of Information Security and Applications, vol. 76, 103532, (2023). DOI:10.1016/j.jisa.2023.103532(CrossRef)(Google Scholar)
[15] V. Hnamte, A. A. Najar, C. Laldinsanga, J. Hussain, and L. Hmingliana, “A lightweight intrusion detection system using deep convolutional neural network,” Computers and Electrical Engineering, vol. 127, 110561, (2025). DOI:10.1016/j.compeleceng.2025.110561(CrossRef)(Google Scholar)
[16] T. H. Trong and T. N. Hoang, “Effective multi-stage training model for edge computing devices in intrusion detection,” International Journal of Computer Networks & Communications, (2024). DOI:10.5121/ijcnc.2024.16102(CrossRef)(Google Scholar)
[17] M. J. Alfadhil, A. Baydoun, M. Alazab, H. U. Rehman, J. A. Jaam, and S. A. S. Al-Maadeed, “Enhancing federated learning for IoT-based anomaly detection: A reputation-based client selection approach,” Alexandria Engineering Journal, vol. 130, pp. 889–909, (2025). DOI:10.1016/j.aej.2025.09.019(CrossRef)(Google Scholar)
[18] C. P. Kaliappan, K. Palaniappan, D. Ananthavadivel, and U. Subramanian, “Advancing IoT security: A comprehensive AI-based trust framework for intrusion detection,” Peer-to-Peer Networking and Applications, vol. 17, pp. 2737–2757, (2024). DOI:10.1007/s12083-024-01684-0(CrossRef)(Google Scholar)
[19] Y. Alghofaili and M. A. Rassam, “A trust management model for IoT devices and services based on multi-criteria decision-making and deep LSTM technique,” Sensors, vol. 22, no. 2, 634 (2021). DOI:10.3390/s22020634(CrossRef)(Google Scholar)
[20] D. Haputhanthri and A. Wijayasiri, “Short-term traffic forecasting using LSTM-based deep learning models,” Proceedings of the Moratuwa Engineering Research Conference (MERCon), pp. 602–607, (2021). DOI:10.1109/MERCon52712.2021.9525670(CrossRef)(Google Scholar)