Federated learning–based anomaly detection for privacy-preserving in IoT-enabled industrial control systems
DOI:
https://doi.org/10.64497/jssci.203Keywords:
Industrial Control Systems, Federated Learning, Anomaly Detection, IIoT, Edge Computing, Cybersecurity.Abstract
The integration of Industrial Internet of Things (IIoT) technologies into Industrial Control Systems (ICS) has significantly expanded the cyber-attack surface of critical infrastructure. Conventional centralized machine learning approaches for anomaly detection are often incompatible with industrial privacy, latency, and availability constraints. This paper presents a federated learning–based anomaly detection framework that enables collaborative model training across distributed edge devices while preserving data locality. The proposed framework leverages edge-based learning and federated aggregation to achieve timely detection of anomalous process behavior without sharing raw operational data. Experimental evaluation using a simulated water treatment ICS testbed demonstrates improved detection accuracy, reduced detection latency, and substantially lower communication overhead compared to centralized learning approaches. These results confirm that federated learning provides a practical and scalable foundation for privacy-preserving cybersecurity monitoring in IoT-enabled industrial environments.
Downloads
References
[1] K. Kairouz et al., “Advances and Open Problems in Federated Learning,” Proc. IEEE, vol. 109, no. 10, pp. 1655–1668, Oct. 2021.
[2] H. B. McMahan et al., “Communication-efficient learning of deep networks from decentralized data,” Proc. AISTATS, pp. 1273–1282, 2017.
[3] A. Humayed, J. Lin, F. Li, and B. Luo, “Cyber-physical systems security—A survey,” IEEE Internet Things J., vol. 4, no. 6, pp. 1802–1831, Dec. 2017. DOI: https://doi.org/10.1109/JIOT.2017.2703172
[4] R. Mitchell and I.-R. Chen, “A survey of intrusion detection techniques for cyber-physical systems,” ACM Comput. Surv., vol. 46, no. 4, pp. 1–29, 2014. DOI: https://doi.org/10.1145/2542049
[5] S. Li, L. Da Xu, and S. Zhao, “The Internet of Things: a survey,” Inf. Syst. Front., vol. 17, no. 2, pp. 243–259, 2015. DOI: https://doi.org/10.1007/s10796-014-9492-7
[6] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015. DOI: https://doi.org/10.1038/nature14539
[7] M. Conti et al., “Internet of Things security and forensics,” Future Gener. Comput. Syst., vol. 78, pp. 544–546, 2018. DOI: https://doi.org/10.1016/j.future.2017.07.060
[8] P. Ferrari et al., “Evaluation of communication latency in industrial IoT applications,” IEEE Trans. Ind. Informatics, vol. 13, no. 2, pp. 709–719, Apr. 2017.
[9] S. Yin et al., “Data-driven techniques focused on modern industry,” IEEE Trans. Ind. Electron., vol. 62, no. 1, pp. 657–667, Jan. 2015. DOI: https://doi.org/10.1109/TIE.2014.2308133
[10] L. Lu et al., “Federated learning for intrusion detection in industrial IoT,” IEEE Access, vol. 8, pp. 197938–197949, 2020.
[11] L. Bottou, “Large-scale machine learning with stochastic gradient descent,” Proc. COMPSTAT, pp. 177–186, 2010. DOI: https://doi.org/10.1007/978-3-7908-2604-3_16
[12] J. Lin et al., “A survey on internet of things: Architecture, enabling technologies, security and privacy,” IEEE Internet Things J., vol. 4, no. 5, pp. 1125–1142, Oct. 2017. DOI: https://doi.org/10.1109/JIOT.2017.2683200
[13] S. Rinaldi, A. Flammini, M. Pasetti, and E. Sisinni, “On the network latency in industrial IoT applications,” IEEE Trans. Ind. Informatics, vol. 13, no. 2, pp. 710–720, 2017.
[14] M. Zubair, H. B. M. Rais, F. Ullah, and A. A. Suleiman, “Enhancing image reconstruction fidelity in low-dose CT scans: A comprehensive MPS-UNet framework,” in Proc. Int. Conf. Smart Cities, Singapore: Springer Nature Singapore, pp. 59–70, 2024. DOI: https://doi.org/10.1007/978-981-96-5848-0_6
[15] H. Daud, N. S. Haron, N. F. M. Krishnan, S. Y. Tan, M. A. F. A. Wahab, A. A. A. Azhar, A. Z. Zubir, S. B. S. Emirlee, D. S. Metwally, and A. A. Suleiman, “Optimizing AWS cloud resource management: Predicting EC2 instance CPU utilization using LSTM and ARIMA models,” Int. J. Anal. Appl., vol. 23, p. 261, 2025. DOI: https://doi.org/10.28924/2291-8639-23-2025-261
[16] M. Zubair, A. A. Suleiman, A. K. Yousafzai, and T. Alazemi, “Prediction of oxide glass refractive index using a novel deep learning architecture, DualNet (U-Net and ANN) with hybrid loss function,” Opt. Quantum Electron., vol. 57, no. 7, p. 387, 2025. DOI: https://doi.org/10.1007/s11082-025-08300-2
[17] T. Kim et al., “Towards federated learning for anomaly detection in cyber-physical systems,” Sensors, vol. 21, no. 7, pp. 1–18, 2021.
[18] N. Moustafa and J. Slay, “The evaluation of network anomaly detection systems: Statistical analysis of UNSW-NB15 data set,” Inf. Secur. J., vol. 25, no. 1–3, pp. 18–31, 2016. DOI: https://doi.org/10.1080/19393555.2015.1125974
[19] A. A. Suleiman, A. Suleiman, U. A. Abdullahi, and S. A. Suleiman, “Estimation of the case fatality rate of COVID-19 epidemiological data in Nigeria using statistical regression analysis,” Biosaf. Health, vol. 3, no. 1, pp. 4–7, 2021. DOI: https://doi.org/10.1016/j.bsheal.2020.09.003
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Obi Ofuka Princewill, Essien Eyo, Bassey I. Ele
This work is licensed under a Creative Commons Attribution 4.0 International License.
- Abstract 143
- PDF 38
