Trust-Aware Federated Learning with Differential Privacy for Secure AIoT in Critical Infrastructures
DOI:
https://doi.org/10.4108/eetiot.10656Keywords:
federated learning, Differential Privacy, Homomorphic Encryption, Graph Neural Networks, Trust-Aware Aggregation, Critical Infrastructures, AIoTAbstract
Federated learning offers a scalable solution for distributed intelligence in Artificial Intelligence of Things (AIoT) systems, yet privacy leakage, adversarial attacks, and system heterogeneity remain persistent challenges in critical infrastructures such as smart cities, agriculture, and forestry. This paper proposes PriSec- FedGuardNet, a trust-aware federated learning framework that integrates differential privacy, homomorphic secure aggregation, and graph neural network–based trust evaluation to safeguard both data and model updates. The framework preserves sensitive information by perturbing gradients with calibrated noise, encrypts local updates for aggregation without decryption, and assigns trust scores to filter unreliable participants. Experimental validation on ToN-IoT, Bot-IoT, and real-world sensor datasets demonstrates that PriSec-FedGuardNet maintains above 97.3% relative utility under strict privacy budgets, improves anomaly detection F1-scores by up to 18% under poisoning attacks, and reduces device-level energy overheads to less than 12%. Domain-specific evaluations across Indian smart city, agricultural, and forestry deployments further highlight the adaptability and efficiency of the framework. By balancing privacy, security, and utility, PriSec-FedGuardNet establishes a robust paradigm for secure federated learning in AIoT-driven critical infrastructures.
Downloads
References
[1] Tan J., Liang Y., Luong N. C., and Niyato D. (2020) “Toward Smart Security Enhancement of Federated Learning Networks,” IEEE Network, vol. 35, no. 1, p. 340. doi: 10.1109/mnet.011.2000379.
[2] Ma J., Naas S., Sigg S., and Lyu X. (2021) “Privacy-preserving Federated Learning based on Multi-key Homomorphic Encryption,” arXiv, doi: 10.48550/arxiv.2104.06824.
[3] Zhao J. et al. (2025) “The Federation Strikes Back: A Survey of Federated Learning Privacy Attacks, Defenses, Applications, and Policy Landscape,” ACM Computing Surveys. doi: 10.1145/3724113.
[4] Goel S., Tibrewal H., Jain A., Pundhir A., and Singh P. (2025) “Secure Generalization through Stochastic Bidirectional Parameter Updates Using Dual-Gradient Mechanism,” doi: 10.48550/ARXIV.2504.02213.
[5] Dong Y.,Wang Y., Gama M.,Mustafa M., Deconinck G., and Huang X. (2024) “Privacy-Preserving Distributed Learning for Residential Short-Term Load Forecasting,” IEEE Internet of Things Journal, vol. 11, no. 9, p. 16817. doi: 10.1109/jiot.2024.3362587.
[6] Asif H. M., Karim M. A., and Kausar F. (2022) “Federated Learning and its Applications for Security and Communication,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 8. doi: 10.14569/ijacsa.2022.0130838.
[7] Ranaweera K., Neiat A. G., Liu X., Kashyap B., and Pathirana P. N. (2025) “Enhancing Federated Learning Through Secure Cluster-Weighted Client Aggregation,” doi: 10.48550/ARXIV.2503.22971.
[8] Jin W. et al. (2023) “FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System,” arXiv, doi:10.48550/arxiv.2303.10837.
[9] Nguyen K., Khan T., Abdinasibfar H., and Michalas A. (2025) “A Privacy-Centric Approach: Scalable and Secure Federated Learning Enabled by Hybrid Homomorphic Encryption,” doi: 10.48550/ARXIV.2507.14853.
[10] Puppala S. and Sinha K. (2025) “Towards Secure and Efficient Farming using Self-Regulating Heterogeneous Federated Learning in Dynamic Network Conditions,” doi: 10.20944/preprints202504.1508.v1.
[11] Zheng Y., Lai S., Liu Y., Yuan X., Yi X., and Wang C. (2022) “Aggregation Service for Federated Learning: An Efficient, Secure, and More Resilient Realization,” arXiv, doi: 10.48550/arxiv.2202.01971.
[12] Korkmaz A. and Rao P. (2025) “A Selective Homomorphic Encryption Approach for Faster Privacy-Preserving Federated Learning,” arXiv, doi: 10.48550/arxiv.2501.12911.10
[13] Han B., Li B., Qi Y., Jurdak R., Huang K., and Yuen C. (2025) “DP2Guard: A Lightweight and Byzantine-Robust Privacy-Preserving Federated Learning Scheme for Industrial IoT,” doi: 10.48550/ARXIV.2507.16134.
[14] Kalapaaking A. P., Stephanie V., Khalil I., Atiquzzaman M., Yi X., and Almashor M. (2022) “SMPC-Based Federated Learning for 6G-Enabled Internet of Medical Things,” IEEE Network, vol. 36, no. 4, p. 182. doi: 10.1109/mnet.007.2100717.
[15] Korkmaz A. and Rao P. (2025) “A Selective Homomorphic Encryption Approach for Faster Privacy-Preserving Federated Learning,” doi: 10.36227/techrxiv.174120692.21508793/v1.
[16] Prasad K. R., Ghosh S., Cormode G., Mironov I., Yousefpour A., and Stock P. (2022) “Reconciling Security and Communication Efficiency in Federated Learning,” arXiv, doi: 10.48550/arxiv.2207.12779.
[17] Du W., Li M., Wu L., Han Y., Zhou T., and Yang X. (2023) “An efficient and robust privacy-preserving framework for cross-device federated learning,” Complex & Intelligent Systems, vol. 9, no. 5, p. 4923. doi: 10.1007/s40747-023-00978-9.
[18] Bonawitz K. et al. (2017) “Practical Secure Aggregation for Privacy-Preserving Machine Learning,” in Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security, p. 1175. doi: 10.1145/3133956.3133982.
[19] Bégui er C., Andreux M., and Tramel E. W. (2020) “Efficient Sparse Secure Aggregation for Federated Learning,” arXiv, doi: 10.48550/arxiv.2007.14861.
[20] Pan Y. H., Zheng C., He W., Yang J., Li H., and Liming W. (2024) “FedSHE: Privacy preserving and efficient federated learning with adaptive segmented CKKS homomorphic encryption,” Cybersecurity, vol. 7, no. 1. doi: 10.1186/s42400-024-00232-w.
[21] Zhao D. (2022) “CHEM: Efficient Secure Aggregation with Cached Homomorphic Encryption in Federated Machine Learning Systems,” arXiv, doi: 10.48550/arxiv.2212.11475.
[22] Batistatos M. C., de Cola T., Kourtis M., Apostolopoulou V., Xilouris G., and Sagias N. C. (2025) “AGRARIAN: A Hybrid AI-Driven Architecture for Smart Agriculture,” doi: 10.20944/preprints202503.1805.v1.
[23] Sandholm T., Mukherjee S., and Huberman B. A. (2021) “SAFE: Secure Aggregation with Failover and Encryption,” arXiv, doi: 10.48550/arxiv.2108.05475.
[24] Ben-Itzhak Y. et al. (2024) “ScionFL: Efficient and Robust Secure Quantized Aggregation,” p. 490. doi: 10.1109/satml59370.2024.00031.
[25] Zhao D. (2023) “Communication-Efficient Search under Fully Homomorphic Encryption for Federated Machine Learning,” arXiv, doi: 10.48550/arxiv.2308.04648.
[26] Liu Z., Chen S., Ye J., Fan J., Li H., and Li X. (2022) “DHSA: Efficient Doubly Homomorphic Secure Aggregation for Cross-silo Federated Learning,” arXiv, doi: 10.48550/arxiv.2208.07189.
[27] de Laage R., Yuhala P., Wicht F.-X., Felber P., Cachin C., and Schiavoni V. (2025) “Practical Secure Aggregation by Combining Cryptography and Trusted Execution Environments,” p. 152. doi: 10.1145/3701717.3730543.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Ramana Kadiyala, C. V. Lakshmi Narayana, S. China Ramu, Narsaiah Putta, Shyam Sunder Pabboju, B. Ramana Reddy
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open-access article distributed under the terms of the Creative Commons Attribution CC BY 4.0 license, which permits unlimited use, distribution, and reproduction in any medium so long as the original work is properly cited.
