ISSN: 2277-405X
Federated Deep Learning for Privacy-Aware Intrusion Detection in IIoT: Architectures, Challenges, and Future Directions
Paper ID: IJATRD-2026-00010
Keywords:
Keywords:
Abstract:
Abstract
The Industrial Internet of Things (IIoT) has sparked a revolution in manufacturing, energy, and critical infrastructure, bringing a new level of connectivity and operational intelligence to machines and devices. This hyper-connectivity, however, comes with new cybersecurity risks that make the use of traditional centralized intrusion detection systems (IDS) impossible, as there are privacy concerns, communication limitations and regulatory requirements that make this difficult in industrial settings. A new paradigm that strikes the balance between high fidelity threat detection and data locality is known as Federated Deep Learning (FDL), where operational data is not sent between nodes but rather only learning models. This survey presents a thorough and taxonomic survey of federated deep learning architectures, training strategies and privacy mechanisms used for intrusion detection in the context of IIoT, based on 63 seminal publications between 2019 and 2025. We comprehensively explore the design space on five dimensions: (1) federated aggregation architectures (flat, hierarchical, asynchronous, peer-to-peer); (2) deep learning model families (DNN, CNN, RNN/LSTM, Autoencoder, Transformer, GNN); (3) privacy-preserving mechanisms (differential privacy, homomorphic encryption, secure multi-party computation, and their hybrids); (4) resilience strategies against adversarial threats such as Byzantine attacks and gradient inversion; (5) benchmark datasets and evaluation protocols. We highlight eight key open challenges: non-IID data heterogeneity, communication efficiency, model personalization, concept drift, hardware constrained deployment, formal verification, and regulatory compliance. We discuss and summarize existing solutions for each challenge and propose specific future research directions. It is a comprehensive survey that is an essential reference for both researchers and practitioners working towards moving the privacy-preserving frontier of IIoT security to the next level.
How to Cite
Patel, P. (2026, June 20).
Federated Deep Learning for Privacy-Aware Intrusion Detection in IIoT: Architectures, Challenges, and Future Directions.
https://ijatrd.org/en/article/2026-00010
References:
References
[1] Statista Research Department, "Number of IoT and non-IoT active device connections worldwide from 2010 to 2025, with forecasts to 2030," Statista, 2024.
[2] H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, "Communication-efficient learning of deep networks from decentralized data," in Proc. 20th Int. Conf. Artif. Intell. Stat. (AISTATS), 2017, pp. 1273–1282.
[3] S. Nakhodchi, A. Dehghantanha, and H. H. Gharaei Manesh, "Privacy and security in smart and precision farming: A bibliometric analysis," in Handbook of Big Data Privacy, Springer, 2020, pp. 305–318.
[4] S. N. Swamy and S. R. Bhatt, "RT-IoT2022: A real-time IoT network intrusion detection dataset," UCI Machine Learning Repository, 2022.
[5] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.
[6] A. Vaswani et al., "Attention is all you need," in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 30, 2017.
[7] V. Mothukuri, R. M. Parizi, S. Pouriyeh, Y. Huang, A. Dehghantanha, and G. Srivastava, "A survey on security and privacy of federated learning," Future Gener. Comput. Syst., vol. 115, pp. 619–640, 2021.
[8] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, no. 7553, pp. 436–444, 2015.
[9] T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Smola, and V. Smith, "Federated optimization in heterogeneous networks," in Proc. 3rd MLSys Conf., 2020, pp. 429–450.
[10] K. Cho et al., "Learning phrase representations using RNN encoder-decoder for statistical machine translation," in Proc. EMNLP, 2014, pp. 1724–1734.
[11] D. P. Kingma and M. Welling, "Auto-encoding variational Bayes," arXiv preprint arXiv:1312.6114, 2013.
[12] [12] I. Ullah and Q. H. Mahmoud, "A scheme for generating a dataset for anomalous activity detection in IoT networks," in Proc. 33rd Can. Conf. AI, 2020, pp. 508–520.
[13] Y. Zhou et al., "A hybrid CNN-LSTM approach for network intrusion detection in IoT environments," IEEE Internet Things J., vol. 9, no. 11, pp. 8902–8913, 2022.
[14] Z. Wu et al., "A comprehensive study on graph neural networks," IEEE Trans. Neural Netw. Learn. Syst., vol. 32, no. 1, pp. 4–24, 2021.
[15] H. Chen et al., "BERT-IDS: Detecting network intrusions using BERT language model," in Proc. IEEE INFOCOM Workshops, 2022.
[16] J. Zhang, B. Chen, S. Yu, and H. Deng, "PEFL: A privacy-enhanced federated learning scheme for intrusion detection," in Proc. IEEE GLOBECOM, 2021, pp. 1–6.
[17] L. Liu et al., "Federated learning for internet of things: A federated learning framework for on-device anomaly data detection," in Proc. ACM SenSys, 2020.
[18] C. Xie, S. Koyejo, and I. Gupta, "Asynchronous federated optimization," arXiv preprint arXiv:1903.03934, 2019.
[19] T. D. Nguyen et al., "DIOT: A federated self-learning anomaly detection system for IoT," in Proc. IEEE ICDCS, 2019, pp. 756–767.
[20] A. Lalitha, S. Shekhar, T. Javidi, and F. Koushanfar, "Fully decentralized federated learning," in Proc. NeurIPS Workshop Fed. Learn., 2018.
[21] H. Liu et al., "Peer-to-peer federated learning for anomaly detection in IIoT," IEEE Trans. Ind. Inform., vol. 18, no. 8, pp. 5323–5333, 2022.
[22] Y. Lu et al., "Blockchain empowered asynchronous federated learning for secure and privacy-preserving digital twin networks," IEEE Netw., vol. 35, no. 6, pp. 60–67, 2021.
[23] C. Thapa, M. A. P. Chamikara, S. Camtepe, and L. Sun, "SplitFed: When federated learning meets split learning," in Proc. AAAI, 2022, pp. 8485–8493.
[24] C. Dwork and A. Roth, "The algorithmic foundations of differential privacy," Found. Trends Theor. Comput. Sci., vol. 9, no. 3–4, pp. 211–407, 2014.
[25] M. Abadi et al., "Deep learning with differential privacy," in Proc. ACM CCS, 2016, pp. 308–318.
[26] M. Naseri, J. Hayes, and E. De Cristofaro, "Local and central differential privacy for robustness and privacy in federated learning," in Proc. NDSS, 2022.
[27] P. Kairouz et al., "Advances and open problems in federated learning," Found. Trends Mach. Learn., vol. 14, no. 1–2, pp. 1–210, 2021.
[28] C. Zhang et al., "BatchCrypt: Efficient homomorphic encryption for cross-silo federated learning," in Proc. USENIX ATC, 2020, pp. 493–506.
[29] S. Truex et al., "A hybrid approach to privacy-preserving federated learning," in Proc. ACM Workshop AISec, 2019, pp. 1–11.
[30] K. Bonawitz et al., "Practical secure aggregation for privacy-preserving machine learning," in Proc. ACM CCS, 2017, pp. 1175–1191.
[31] T. Lin, L. Kong, S. U. Stich, and M. Jaggi, "Ensemble distillation for robust model fusion in federated learning," in Proc. NeurIPS, 2020.
[32] D. Zhu, J. Hong, and T. Zhang, "Data-free knowledge distillation for heterogeneous federated learning," in Proc. ICML, 2021, pp. 12878–12889.
[33] A. N. Bhagoji, S. Chakraborty, P. Mittal, and S. Calo, "Analyzing federated learning through an adversarial lens," in Proc. ICML, 2019, pp. 634–643.
[34] S. A. Rahman et al., "Federated learning-based intrusion detection system for Internet of Things," in Proc. IEEE GLOBECOM, 2020.
[35] D. Yin, Y. Chen, R. Kannan, and P. Bartlett, "Byzantine-robust distributed learning: Towards optimal statistical rates," in Proc. ICML, 2018, pp. 5650–5659.
[36] E. M. El Mhamdi, R. Guerraoui, and S. Rouault, "The hidden vulnerability of distributed learning in Byzantium," in Proc. ICML, 2018, pp. 3521–3530.
[37] C. Cao, M. Fang, N. Gong, and J. Weng, "FLTrust: Byzantine-robust federated learning via trust bootstrapping," in Proc. NDSS, 2022.
[38] N. Nguyen et al., "FLAME: Taming backdoors in federated learning," in Proc. USENIX Security, 2022, pp. 1255–1272.
[39] L. Zhu, Z. Liu, and S. Han, "Deep leakage from gradients," in Proc. NeurIPS, 2019, pp. 14774–14784.
[40] O. Zari, C. Xu, and J. S. Giraldo, "Efficient passive membership inference attack in federated learning," arXiv preprint arXiv:2111.00430, 2021.
[41] V. Rey, P. M. Sanchez Sanchez, A. H. Celdran, and G. Bovet, "Federated learning for malware detection in IoT devices," Comput. Netw., vol. 204, p. 108693, 2022.
[42] O. Friha, M. A. Ferrag, L. Shu, L. Maglaras, and X. Wang, "FELIDS: Federated learning-based intrusion detection system for agricultural IoT," J. Parallel Distrib. Comput., vol. 165, pp. 17–31, 2022.
[43] S. I. Popoola et al., "Federated deep learning for zero-day botnet attack detection in IoT edge devices," IEEE Internet Things J., vol. 9, no. 5, pp. 3930–3944, 2022.
[44] T. D. Nguyen et al., "FLAME: Differentially private federated learning in the shuffle model," in Proc. AAAI, 2022, pp. 9387–9395.
[45] J. Zhao et al., "Federated learning with non-IID data," arXiv preprint arXiv:1806.00582, 2022.
[46] S. Han, H. Kim, D. Kwon, and K. Lee, "Communication-efficient federated learning for IoT intrusion detection with gradient compression," IEEE Access, vol. 11, pp. 47621–47635, 2023.
[47] B. Ghimire and D. B. Rawat, "Recent advances on federated learning for cybersecurity and cybersecurity for federated learning for Internet of Things," IEEE Internet Things J., vol. 9, no. 11, pp. 8229–8249, 2022.
[48] M. A. Ferrag et al., "Edge-IIoTset: A new comprehensive realistic cyber security dataset," IEEE Access, vol. 10, pp. 40281–40306, 2022.
[49] L. Cui et al., "Security and privacy-enhanced federated learning for anomaly detection in IoT infrastructures," IEEE Trans. Ind. Inform., vol. 18, no. 5, pp. 3374–3384, 2023.
[50] K. Hsieh et al., "Quagmire: Non-IID problems in federated learning," in Proc. NeurIPS FL Workshop, 2020.
[51] D. Preuveneers et al., "Chained anomaly detection models for federated learning," Appl. Sci., vol. 8, no. 12, p. 2663, 2021.
[52] S. Karimireddy et al., "SCAFFOLD: Stochastic controlled averaging for federated learning," in Proc. ICML, 2020, pp. 5132–5143.
[53] Q. Li et al., "Model-contrastive federated learning," in Proc. CVPR, 2021, pp. 10713–10722.
[54] J. Goodfellow et al., "Generative adversarial nets," in Proc. NeurIPS, 2014, pp. 2672–2680.
[55] C. Guha, A. Talwalkar, and V. Smith, "One-shot federated learning," arXiv preprint arXiv:1902.11175, 2019.
[56] J. Kirkpatrick et al., "Overcoming catastrophic forgetting in neural networks," Proc. Natl. Acad. Sci., vol. 114, no. 13, pp. 3521–3526, 2017.
[57] A. Madry, A. Makelov, L. Schmidt, D. Tsipras, and A. Vladu, "Towards deep learning models resistant to adversarial attacks," in Proc. ICLR, 2018.
[58] C. Finn, P. Abbeel, and S. Levine, "Model-agnostic meta-learning for fast adaptation of deep networks," in Proc. ICML, 2017, pp. 1126–1135.
[59] T. Chen et al., "A simple framework for contrastive learning of visual representations," in Proc. ICML, 2020, pp. 1597–1607.
[60] P. Veličković et al., "Graph attention networks," in Proc. ICLR, 2018.
[61] B. Li, Y. Wu, J. Song, R. Lu, T. Li, and L. Zhao, "DeepFed: Federated deep learning for intrusion detection in industrial cyber-physical systems," IEEE Trans. Ind. Informat., vol. 17, no. 8, pp. 5615–5624, Aug. 2021.
[62] S. Sharma, K. Bellare, and P. Mehta, "Privacy-preserving hierarchical fog federated learning (PP-HFFL) for IoT intrusion detection," Sensors, vol. 25, no. 22, p. 7296, 2025.
[63] Y. Zhang, M. Chen, and X. Wang, "Toward enhancing privacy preservation of a federated learning CNN intrusion detection system in IoT: Method and empirical study," ACM Trans. Softw. Eng. Methodol., vol. 34, no. 1, pp. 1–35, Jan. 2025.
