Research Article
Trust Bounded Federated Anomaly Detection for Industrial IoT Networks
Abstract
Federated anomaly detection can keep industrial traffic at its originating site, yet locality alone does not determine who can inspect model updates, what a privacy budget protects, or whether an alarm controller observes valid feedback. This paper recasts a proposed Industrial Internet of Things (IIoT) detector as four linked contracts: local temporal inference, encrypted linear aggregation, differentially private release, and bounded threshold actuation. The distinction matters because each contract has a different adversary, parameter record, and validation burden. The source reports 93.3% accuracy, 90.2% precision, 89.8% recall, and 90.1% on X-IIoTID, together with 9.2 s and 5.2 J per round. Those values are retained as author-reported observations rather than reproduced findings. The supplied record does not identify the evaluated subset, client partition, executable implementation, cryptographic parameters, privacy accountant, raw predictions, hardware, or measurement traces; its dataset summary also lacks a documented mapping to the released X-IIoTID corpus. The resulting contribution is therefore an auditable architecture and evidence contract, not a verified performance claim. Publication-level validation requires a leakage-safe data identity, matched baselines, explicit threat and adjacency models, seed-level predictions, and end-to-end resource accounting.
Keywords
Industrial internet of things Federated anomaly detection Trust boundary Homomorphic aggregation Differential privacy Adaptive thresholdingReferences (44)
- [1] McMahan, B., Moore, E., Ramage, D., Hampson, S., & y Arcas, B. A. (2017). Communication-efficient learning of deep networks from decentralized data. Artificial Intelligence and Statistics (pp. 1273-1282). Pmlr. https://doi.org/10.48550/arXiv.1602.05629
- [2] Kairouz, P., & McMahan, H. B. (2021). Advances and open problems in federated learning. Foundations and Trends in Machine Learning, 14(1–2), 1–210. https://doi.org/10.1561/2200000083
- [3] Geiping, J., Bauermeister, H., Dröge, H., & Moeller, M. (2020). Inverting gradients-how easy is it to break privacy in federated learning? Advances in Neural Information Processing Systems, 33, 16937–16947. https://doi.org/10.48550/arXiv.2003.14053
- [4] Li, T., Sahu, A. K., Zaheer, M., Sanjabi, M., Talwalkar, A., & Smith, V. (2020). Federated optimization in heterogeneous networks. Proceedings of Machine Learning and Systems, 2, 429–450. https://doi.org/10.48550/arXiv.1812.06127
- [5] Li, B., Wu, Y., Song, J., Lu, R., Li, T., & Zhao, L. (2020). DeepFed: Federated deep learning for intrusion detection in industrial cyber--physical systems. IEEE Transactions on Industrial Informatics, 17(8), 5615–5624. https://doi.org/10.1109/TII.2020.3023430
- [6] He, N., Zhang, Z., Wang, X., & Gao, T. (2023). Efficient privacy-preserving federated deep learning for network intrusion of industrial IoT. International Journal of Intelligent Systems, 2023(1), 2956990. https://doi.org/10.1155/2023/2956990
- [7] Alqazzaz, A. (2026). SecuFL-IoT: An adaptive privacy-preserving federated learning framework for anomaly detection in smart industrial networks. Scientific Reports, 16(1), 4107. https://doi.org/10.1038/s41598-025-11883-1%0A%0A
- [8] Dwork, C., McSherry, F., Nissim, K., & Smith, A. (2006). Calibrating noise to sensitivity in private data analysis. Theory of Cryptography Conference (pp. 265-284). Berlin, Heidelberg: Springer Berlin Heidelberg.. https://doi.org/10.1007/11681878_14%0A%0A
- [9] Cheon, J. H., Kim, A., Kim, M., & Song, Y. (2017). Homomorphic encryption for arithmetic of approximate numbers. International Conference on the Theory and Application of Cryptology and Information Security (pp. 409-437). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-319-70694-8_15
- [10] Karimireddy, S. P., Kale, S., Mohri, M., Reddi, S., Stich, S., & Suresh, A. T. (2020). Scaffold: Stochastic controlled averaging for federated learning. International Conference on Machine Learning (pp. 5132-5143). PMLR. https://proceedings.mlr.press/v119/karimireddy20a.html
- [11] Friha, O., Ferrag, M. A., Benbouzid, M., Berghout, T., Kantarci, B., & Choo, K.K. R. (2023). 2DF-IDS: Decentralized and differentially private federated learning-based intrusion detection system for industrial IoT. Computers & Security, 127, 103097. https://doi.org/10.1016/j.cose.2023.103097
- [12] Arp, D., Quiring, E., Pendlebury, F., Warnecke, A., Pierazzi, F., Wressnegger, C., ..., & Rieck, K. (2022). Dos and don'ts of machine learning in computer security. 31st USENIX security symposium (pp. 3971-3988). USENIX Security 22. https://www.usenix.org/system/files/sec22summer_arp.pdf
- [13] Liu, Q., & Paparrizos, J. (2024). The elephant in the room: Towards a reliable time-series anomaly detection benchmark. Advances in Neural Information Processing Systems, 37, 108231–108261. https://doi.org/10.52202/079017-3437
- [14] Wang, Y., Wang, X., Yi, B., Huang, M., & Kou, Y. (2025). P3Fed: A personalized and privacy-preserving federated framework for intrusion detection in computing power network. Computer Networks, 111886. https://doi.org/10.1016/j.comnet.2025.111886
- [15] Gu, X., Li, M., & Xiong, L. (2025). {DP-BREM}:{Differentially-Private} and {Byzantine-Robust} federated learning with client momentum. 34th USENIX Security Symposium (pp. 3065-3082). USENIX. https://www.usenix.org/conference/usenixsecurity25/presentation/gu-xiaolan
- [16] Garg, S., Kavousi, A., Kolonelos, D., Tairi, E., & Wang, Z. (2026). Lighthouse: Single-server secure aggregation with $ o (1) $ server-committee communication at scale. Cryptology Eprint Archive. https://ia.cr/2026/346
- [17] Alencar, B., Cassales, G., Gomes, H. M., Prazeres, C., Rios, T., Bifet, A., & Rios, R. (2026). Concept drift detection in delayed and partially labeled data streams: An experimental survey. Digital Signal Processing, 106325. https://doi.org/10.1016/j.dsp.2026.106325
- [18] Miller, T., & Durlik, I. (2026). Can federated learning go green? EcoFL: A system-level energy-aware benchmark for IoT edge intelligence. Journal of Low Power Electronics and Applications, 16(3), 24. https://doi.org/10.3390/jlpea16030024
- [19] Wang, X., Wang, Y., Javaheri, Z., Almutairi, L., Moghadamnejad, N., & Younes, O. S. (2023). Federated deep learning for anomaly detection in the internet of things. Computers and Electrical Engineering, 108, 108651. https://doi.org/10.1016/j.compeleceng.2023.108651
- [20] Sun, S., Zhou, L., Wang, Z., & Han, L. (2025). Robust intrusion detection based on personalized federated learning for IoT environment. Computers & Security, 154, 104442. https://doi.org/10.1016/j.cose.2025.104442
- [21] Hu, T., Chen, Q., Hu, Y., Hou, S., Yan, H., Yi, P., & Cui, Z. (2025). Efficient and privacy-preserving network intrusion detection based on federated learning in sdn-enabled iiot network. IEEE Internet of Things Journal, 12(20), 41904–41923. https://doi.org/10.1109/JIOT.2025.3591598
- [22] Zhang, K., Huang, R., Wang, J., Wang, X., Wang, W., Yan, Y., & Shi, X. (2026). CC-FIDS: A federated intrusion detection system with low false negative rates via category consensus and hierarchical filtering. Journal of King Saud University Computer and Information Sciences, 38(7), 651. https://doi.org/10.1007/s44443-026-01045-2
- [23] Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., ..., & Seth, K. (2017). Practical secure aggregation for privacy-preserving machine learning. Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security (pp. 1175-1191). ACM. https://doi.org/10.1145/3133956.3133982
- [24] Xu, C., Du, X., Li, L., Li, X., & Yu, H. (2024). End-edge collaborative lightweight secure federated learning for anomaly detection of wireless industrial control systems. IEEE Open Journal of the Industrial Electronics Society, 5, 132–142. https://doi.org/10.1109/OJIES.2024.3370496
- [25] Mouchet, C., Troncoso-Pastoriza, J., Bossuat, J. P., & Hubaux, J. P. (2021). Multiparty homomorphic encryption from ring-learning-with-errors. Proceedings on Privacy Enhancing Technologies, 2021(4), 291–311. https://doi.org/10.2478/popets-2021-0071
- [26] Qin, X., Yang, X., & Tang, X. (2026). Practical privacy-preserving federated learning based on multiparty homomorphic encryption for large-scale models. Pattern Recognition, 171, 112174. https://doi.org/10.1016/j.patcog.2025.112174
- [27] Liu, Y., Jia, Z., Chen, X., Bian, S., Xu, R., Li, D., & Lu, Y. (2025). Aion: Robust and efficient {multi-round}{single-mask} secure aggregation against malicious participants. 34th USENIX security symposium (pp. 3025-3044). USENIX. https://www.usenix.org/conference/usenixsecurity25/presentation/liu-yizhong
- [28] Pasquini, D., Francati, D., & Ateniese, G. (2022, November). Eluding secure aggregation in federated learning via model inconsistency. Proceedings of the 2022 ACM SIGSAC conference on computer and communications security (pp. 2429-2443). ACM. https://doi.org/10.1145/3548606.3560557
- [29] Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016). Deep learning with differential privacy. Proceedings of the 2016 ACM SIGSAC conference on computer and communications security (pp. 308-318). ACM. https://doi.org/10.1145/2976749.2978318
- [30] Mironov, I. (2017, August). Rényi differential privacy. 2017 IEEE 30th computer security foundations symposium (CSF) (pp. 263-275). IEEE. https://doi.org/10.1109/CSF.2017.11
- [31] Wang, Y. X., Balle, B., & Kasiviswanathan, S. P. (2019). Subsampled rényi differential privacy and analytical moments accountant. The 22nd International Conference on Artificial Intelligence and Statistics (pp. 1226-1235). PMLR. https://proceedings.mlr.press/v89/wang19b.html
- [32] Gopi, S., Lee, Y. T., & Wutschitz, L. (2021). Numerical composition of differential privacy. Advances in Neural Information Processing Systems, 34, 11631–11642. https://doi.org/10.29012/jpc.870
- [33] Al-Hawawreh, M., Sitnikova, E., & Aboutorab, N. (2021). X-IIoTID: A connectivity-agnostic and device-agnostic intrusion data set for industrial Internet of Things. IEEE Internet of Things Journal, 9(5), 3962–3977. https://doi.org/10.1109/JIOT.2021.3102056
- [34] Al-Hawawreh, M., Sitnikova, E., & Aboutorab, N. (2021). X-IIoTID: A connectivity-agnostic and device-agnostic intrusion data set for industrial internet of things. IEEE Internet of Things Journal, 9(5), 3962-3977. https://doi.org/10.1109/JIOT.2021.3102056
- [35] Kapoor, S., & Narayanan, A. (2023). Leakage and the reproducibility crisis in machine-learning-based science. Patterns, 4(9). https://doi.org/10.1016/j.patter.2023.100804
- [36] Chawla, N. V, Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. https://doi.org/10.1613/jair.953
- [37] Kim, S., Choi, K., Choi, H. S., Lee, B., & Yoon, S. (2022). Towards a rigorous evaluation of time-series anomaly detection. In Proceedings of the AAAI conference on artificial intelligence (pp. 7194-7201). AAAI. https://doi.org/10.1609/aaai.v36i7.20680
- [38] Wagner, D., Nair, A., Franks, B. J., Arweiler, J., Muraleedharan, A., Jungjohann, I., ..., & Kloft, M. (2025). Formally exploring time-series anomaly detection evaluation metrics. https://doi.org/10.48550/arXiv.2510.17562
- [39] Zhong, Z., Song, L., Tang, F., & Yuan, B. (2024). Addressing intermediate verification latency in online learning through immediate pseudo-labeling and oriented synthetic correction. 2024 International Joint Conference on Neural Networks (IJCNN) (pp. 1-10). IEEE. https://doi.org/10.1109/IJCNN60899.2024.10650783
- [40] Gibbs, I., & Candès, E. J. (2024). Conformal inference for online prediction with arbitrary distribution shifts. Journal of Machine Learning Research, 25(162), 1-36. https://doi.org/10.48550/arXiv.2208.08401
- [41] Siffer, A., Fouque, P. A., Termier, A., & Largouet, C. (2017). Anomaly detection in streams with extreme value theory. Proceedings of the 23rd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1067-1075). ACM. https://doi.org/10.1145/3097983.3098144
- [42] Shin, J. S. (2025). SAFE-IDS: A privacy-preserving framework for overcoming non-IID challenges in federated intrusion detection. Computers & Security, 155, 104492. https://doi.org/10.1016/j.cose.2025.104492
- [43] Lai, F., Dai, Y., Singapuram, S., Liu, J., Zhu, X., Madhyastha, H., & Chowdhury, M. (2022). Fedscale: Benchmarking model and system performance of federated learning at scale. International Conference on Machine Learning (pp. 11814-11827). PMLR. https://proceedings.mlr.press/v162/lai22a.html
- [44] Henderson, P., Hu, J., Romoff, J., Brunskill, E., Jurafsky, D., & Pineau, J. (2020). Towards the systematic reporting of the energy and carbon footprints of machine learning. Journal of Machine Learning Research, 21(248), 1–43. https://jmlr.org/papers/v21/20-312.html