Fed-IoT-Sec: Privacy-Preserving Federated Autoencoder for Anomaly Detection in IoT Networks
DOI:
https://doi.org/10.21070/jicte.v10i1.1709Keywords:
Internet Of Things, Federated Learning, Anomaly Detection, Autoencoder, Privacy PreservationAbstract
General Background: The rapid expansion of Internet of Things (IoT) networks has introduced significant cybersecurity challenges due to resource-constrained devices and decentralized infrastructures that complicate conventional protection mechanisms. Specific Background: Many existing anomaly detection approaches rely on centralized data processing, which can create privacy risks and communication bottlenecks in distributed IoT environments. Knowledge Gap: Current solutions often fail to simultaneously address privacy preservation, communication efficiency, and the heterogeneous non-IID data characteristics commonly observed across IoT devices. Aims: This study proposes Fed-IoT-Sec, a lightweight privacy-preserving federated learning framework that integrates an autoencoder-based anomaly detection model with a federated training protocol suitable for resource-limited IoT systems. Results: Experimental evaluation using the NSL-KDD dataset demonstrates that the proposed framework achieves a detection accuracy of 96.4%, reaching performance close to centralized autoencoder models while reducing communication overhead by 30% compared with traditional federated learning approaches. Novelty: The framework combines a compact autoencoder architecture with a federated averaging protocol designed to accommodate non-IID data distributions and maintain data locality without transmitting raw device information. Implications: These findings indicate that the proposed approach provides a practical and secure anomaly detection mechanism for IoT environments, supporting collaborative model training while preserving privacy and reducing network communication costs.
Highlights:
• Lightweight Federated Architecture Designed for Resource-Limited Connected Devices
• Collaborative Model Training Without Transferring Raw Network Traffic Records
• Experimental Evaluation Reports 96.4% Detection Accuracy and 30% Bandwidth Reduction
Keywords: Internet Of Things, Federated Learning, Anomaly Detection, Autoencoder, Privacy Preservation.
References
[1] J. Lin et al., "A Survey of Industrial AIoT: Opportunities, Challenges, and Directions," IEEE Access, vol. 12, pp. 102450-102475, 2024, doi: 10.1109/ACCESS.2024.1059198.
[2] S. Khan and A. Gani, "Comprehensive Study of IoT Vulnerabilities and Countermeasures," Appl. Sci., vol. 15, no. 6, p. 3036, 2025, doi: 10.3390/app15063036.
[3] M. A. Ferrag et al., "A Survey of Cross-Layer Security for Resource-Constrained IoT Devices," MDPI Electronics, vol. 14, no. 2, p. 412, 2025, doi: 10.3390/electronics14020412.
[4] T. Li, A. K. Sahu, A. Talwalkar and V. Smith, "Federated Learning: Challenges, Methods, and Future Directions," IEEE Signal Process. Mag., vol. 37, no. 3, pp. 50-60, May 2020, doi: 10.1109/MSP.2020.2975807.
[5] R. Chien et al., "Federated Learning-Based Intrusion Detection in IoT Networks: Performance Evaluation and Data Scaling Study," Sensors, vol. 25, no. 4, p. 1152, 2025, doi: 10.3390/s25041152.
[6] Q. Li et al., "A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection," IEEE Trans. Knowl. Data Eng., vol. 35, no. 4, pp. 3347-3366, April 2023, doi: 10.1109/TKDE.2021.3124599.
[7] S. L. Mohammed et al., "Communication-Efficient Federated Learning for Wireless Edge Intelligence in IoT," IEEE Internet Things J., vol. 7, no. 5, pp. 4390-4401, May 2020, doi: 10.1109/JIOT.2020.2971231.
[8] H. Zhang and G. Sun, "Robust Federated Learning Against Data Poisoning Attacks: Prevention and Detection of Attacked Nodes," Electronics, vol. 14, no. 15, p. 2970, 2025, doi: 10.3390/electronics14152970.
[9] Y. Chen et al., "Enhancing Anomaly Detection in Distributed Power Systems Using Autoencoder-Based Federated Learning," Front. Energy Res., vol. 11, p. 104378, 2023, doi: 10.3389/fenrg.2023.10437833.
[10] V. Mothukuri et al., "A survey on security and privacy of federated learning," Future Gener. Comput. Syst., vol. 115, pp. 619-640, Feb. 2021, doi: 10.1016/j.future.2020.10.007.
[11] T. Li et al., "Federated Learning: Challenges, Methods, and Future Directions," IEEE Signal Process. Mag., vol. 37, no. 3, pp. 50-60, May 2020, doi: 10.1109/MSP.2020.2975807.
[12] D. C. Nguyen et al., "Federated Learning for Internet of Things: A Comprehensive Survey," IEEE Commun. Surveys Tuts., vol. 23, no. 3, pp. 1622-1658, 2021, doi: 10.1109/COMST.2021.3075439.
[13] Z. Zhang et al., "A Survey on Federated Learning in the Era of 6G Communications," IEEE Commun. Surveys Tuts., vol. 26, no. 1, pp. 1-35, 2024, doi: 10.1109/COMST.2023.3323450.
[14] M. A. Ferrag et al., "Federated Deep Learning for Cyber Security in the Internet of Things: Concepts, Applications, and Challenges," IEEE Network, vol. 35, no. 5, pp. 270-277, 2021, doi: 10.1109/MNET.011.2000709.
[15] R. Saha et al., "A Survey on Federated Learning: From Theory to Practice in Edge Computing," IEEE Access, vol. 10, pp. 10580-10605, 2022, doi: 10.1109/ACCESS.2022.3144185.
[16] S. Taneja et al., "Deep Learning-based Anomaly Detection in IoT: A Survey and Taxonomy," Journal of Network and Systems Management, vol. 33, no. 1, p. 12, 2025, doi: 10.1007/s10922-024-09812-4.
[17] Y. Otoum and A. Nayak, "AS-IDS: Anomaly-Based Smart Intrusion Detection System for IoT," IEEE Internet Things J., vol. 8, no. 16, pp. 12908-12917, 2021, doi: 10.1109/JIOT.2021.3064886.
[18] G. Pang et al., "Deep Learning for Anomaly Detection: A Review," ACM Comput. Surv., vol. 54, no. 2, pp. 1-38, 2021, doi: 10.1145/3439950.
[19] L. Yang et al., "Lightweight Autoencoder-based Federated Learning for IoT Malware Detection," IEEE Trans. Inf. Forensics Security, vol. 19, pp. 1245-1258, 2024, doi: 10.1109/TIFS.2023.3345120.
[20] A. Derhab et al., "Blockchain and Random Forest-Based IDS for Securing SDN-Enabled Industrial IoT," IEEE Trans. Ind. Informat., vol. 16, no. 11, pp. 7150-7159, 2020, doi: 10.1109/TII.2020.2972342.
[21] R. Vinayakumar et al., "Deep Learning for Intelligent Healthcare Systems: A Review," IEEE Access, vol. 11, pp. 1583-1610, 2023, doi: 10.1109/ACCESS.2023.3234501.
[22] A. D. Diro and N. Chilamkurti, "Distributed Attack Detection Scheme using Deep Learning Approach for Internet of Things," Future Gener. Comput. Syst., vol. 115, pp. 619-630, Feb. 2021, doi: 10.1016/j.future.2020.10.007.
[23] Y. Liu et al., "A Survey on Federated Learning: Architecture and Strategy," IEEE Trans. Neural Netw. Learn. Syst., vol. 35, no. 3, pp. 3125-3142, 2024, doi: 10.1109/TNNLS.2022.3212567.
[24] T. Zhang et al., "Anomaly Detection in IoT Systems via Federated Learning," IEEE Internet Things J., vol. 10, no. 5, pp. 4500-4512, 2023, doi: 10.1109/JIOT.2022.3211100.
[25] Y. Chen et al., "Privacy-Preserving Federated Learning for Industrial IoT Anomaly Detection," IEEE Trans. Ind. Informat., vol. 20, no. 2, pp. 1540-1552, 2024, doi: 10.1109/TII.2023.3289012.
[26] Y. Zhao et al., "Federated Learning with Non-IID Data," IEEE Trans. Knowl. Data Eng., vol. 36, no. 4, pp. 1820-1833, 2024, doi: 10.1109/TKDE.2023.3290110.
[27] V. Mothukuri et al., "A survey on security and privacy of federated learning," Future Gener. Comput. Syst., vol. 115, pp. 619-640, 2021, doi: 10.1016/j.future.2020.10.007.
[28] D. C. Nguyen et al., "Federated Learning for Smart Healthcare: A Survey," ACM Comput. Surv., vol. 55, no. 3, pp. 1-37, 2022, doi: 10.1145/3501296.
[29] M. A. Ferrag et al., "Federated Deep Learning for Cyber Security in the Internet of Things," IEEE Network, vol. 35, no. 5, pp. 270-277, 2021, doi: 10.1109/MNET.011.2000709.
[30] M. Al-Kasasbeh et al., "A Lightweight Intrusion Detection System for IoT Networks Using Machine Learning," Sensors, vol. 25, no. 1, p. 112, 2025, doi: 10.3390/s25010112.
Downloads
Published
License
Copyright (c) 2026 Raad Qasim
This work is licensed under a Creative Commons Attribution 4.0 International License.
