Privacy-Preserving Asynchronous Federated Learning for Heterogeneous IoT Devices

Qiao Xue

College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China

*Corresponding author

Federated learning (FL), as a distributed machine learning paradigm, is particularly suitable for Internet of Things (IoT) scenarios where numerous edge devices collaboratively train a global model without sharing raw data, thereby reducing the risk of privacy leakage. However, IoT environments are typically characterized by device heterogeneity, unstable network connections, and limited computational and communication resources. These factors pose significant challenges to both training efficiency and privacy protection. For instance, resource-constrained IoT devices may slow down overall training progress, leading to inefficient resource utilization. Moreover, the gradient information uploaded by IoT devices can still be exploited by malicious attackers, resulting in potential privacy breaches. To address these issues, this thesis proposes a secure asynchronous federated learning algorithm tailored for IoT device-heterogeneous environments. The algorithm leverages model pruning to allocate sub-models with varying complexities according to the computational and communication capabilities of IoT clients, thereby improving resource utilization and accelerating training on low-end devices. Furthermore, it incorporates a staleness-aware asynchronous aggregation mechanism, dynamically adjusting aggregation weights to mitigate the negative effects of stale updates from delayed devices on the global model. To further enhance privacy protection, a differential privacy mechanism is integrated into the local training process by injecting carefully calibrated noise into gradient information, effectively preventing sensitive data leakage. Experimental results demonstrate that the proposed algorithm achieves superior comprehensive performance in model accuracy, convergence speed, and privacy protection strength, outperforming baseline asynchronous federated learning algorithms in device-heterogeneous environments.

Asynchronous federated learning, Internet of Things, Differential privacy, Device heterogeneity, Privacy security

Hui Cheng, Qiao Xue, Youwen Zhu (2025). Privacy-Preserving Asynchronous Federated Learning for Heterogeneous IoT Devices. Journal of Networking and Network Applications, Volume 5, Issue 1, pp. 27–38. https://doi.org/10.33969/J-NaNA.2025.050103.

[1] Z. Ghahramani, “Probabilistic machine learning and artificial intelli-gence,” Nature, vol. 521, no. 7553, pp. 452–459, 2015.

[2] C. Zhang, Y. Xie, H. Bai, B. Yu, W. Li, and Y. Gao, “A survey on federated learning,” Knowledge-Based Systems, vol. 216, p. 106775, 2021.

[3] B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Artificial intelligence and statistics. PMLR, 2017, pp. 1273–1282.

[4] C. Xu, Y. Qu, Y. Xiang, and L. Gao, “Asynchronous federated learning on heterogeneous devices: A survey,” Computer Science Review, vol. 50, p. 100595, 2023.

[5] F. D. Protection, “General data protection regulation (gdpr),” Intersoft Consulting, Accessed in October, vol. 24, no. 1, 2018.

[6] M. Nasr, R. Shokri, and A. Houmansadr, “Comprehensive privacy analysis of deep learning: Passive and active white-box inference attacks against centralized and federated learning,” in 2019 IEEE symposium on security and privacy (SP). IEEE, 2019, pp. 739–753.

[7] R. C. Geyer, T. Klein, and M. Nabi, “Differentially private federated learning: A client level perspective,” arXiv preprint arXiv:1712.07557, 2017.

[8] C. Dwork, F. McSherry, K. Nissim, and A. Smith, “Calibrating noise to sensitivity in private data analysis,” in Theory of cryptography conference. Springer, 2006, pp. 265–284.

[9] C. Dwork, “Differential privacy,” in International colloquium on au-tomata, languages, and programming. Springer, 2006, pp. 1–12.

[10] C. Dwork, K. Kenthapadi, F. McSherry, I. Mironov, and M. Naor, “Our data, ourselves: Privacy via distributed noise generation,” in Annual international conference on the theory and applications of cryptographic techniques. Springer, 2006, pp. 486–503.

[11] E. Diao, J. Ding, and V. Tarokh, “Heterofl: Computation and commu-nication efficient federated learning for heterogeneous clients,” arXiv preprint arXiv:2010.01264, 2020.

[12] T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V. Smith, “Federated optimization in heterogeneous networks,” Proceedings of Machine learning and systems, vol. 2, pp. 429–450, 2020.

[13] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, 2002.

[14] A. Krizhevsky, G. Hinton et al., “Learning multiple layers of features from tiny images.(2009),” 2009.

[15] C. Xie, S. Koyejo, and I. Gupta, “Asynchronous federated optimization,” arXiv preprint arXiv:1903.03934, 2019.

[16] E. Jeong, S. Oh, H. Kim, J. Park, M. Bennis, and S.-L. Kim, “Communication-efficient on-device machine learning: Federated dis-tillation and augmentation under non-iid private data,” arXiv preprint arXiv:1811.11479, 2018.

[17] W. Wu, L. He, W. Lin, R. Mao, C. Maple, and S. Jarvis, “Safa: A semi-asynchronous protocol for fast federated learning with low overhead,” IEEE Transactions on Computers, vol. 70, no. 5, pp. 655–668, 2020.

[18] J. Nguyen, K. Malik, H. Zhan, A. Yousefpour, M. Rabbat, M. Malek, and D. Huba, “Federated learning with buffered asynchronous aggregation,” in International conference on artificial intelligence and statistics. PMLR, 2022, pp. 3581–3607.

[19] J. Ziegler, B. Pfitzner, H. Schulz, A. Saalbach, and B. Arnrich, “Defend-ing against reconstruction attacks through differentially private federated learning for classification of heterogeneous chest x-ray data,” Sensors, vol. 22, no. 14, p. 5195, 2022.

[20] A. El Ouadrhiri and A. Abdelhadi, “Differential privacy for deep and federated learning: A survey,” IEEE access, vol. 10, pp. 22 359–22 380, 2022.

[21] G. Yang, S. Wang, and H. Wang, “Federated learning with personalized local differential privacy,” in 2021 IEEE 6th International Conference on Computer and Communication Systems (ICCCS). IEEE, 2021, pp. 484–489.

[22] X. Lu, Y. Liao, P. Lio, and P. Hui, “Privacy-preserving asynchronous federated learning mechanism for edge network computing,” Ieee Ac-cess, vol. 8, pp. 48 970–48 981, 2020.

[23] B. Li, H. Gao, and X. Shi, “Feddp: Secure federated learning for disease prediction with imbalanced genetic data,” bioRxiv, pp. 2023–01, 2023.