Over-the-Air Computation Enabled Semi-Asynchronous Wireless Federated Learning
IEEE Transactions on Communications · 2025-10 · 73 (10) · 8919-8936
DOI: 10.1109/tcomm.2025.3582727 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Communications, 2025-10. Verfasst von Zijian Zheng und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: Engineering and Physical Sciences Research Council (EPSRC), U.K..
Abstract (Originalsprache)
The emerging field of federated learning (FL) holds significant promise for advancing edge intelligence while preserving data privacy. However, as FL systems scale or become more heterogeneous, challenges such as spectrum scarcity and the straggler problem arise. To address these issues, this paper proposes SA-AirFed, a semi-asynchronous FL architecture compatible with Over-the-Air Computation (AirComp). We develop an efficient scheduling scheme that meets AirComp’s requirements and analyze the factors affecting convergence under the Lipschitz-Smooth condition. Building on insights from the convergence analysis, we design an adaptive algorithm that mitigates staleness from semi-asynchronous aggregation and noise from AirComp by dynamically adjusting aggregation weights, formulated as a convex quadratic programming problem. Experimental results on MNIST and CIFAR-10 demonstrate that SA-AirFed significantly reduces wall-clock training time while achieving greater robustness compared to baseline models.
Förderung
- Engineering and Physical Sciences Research Council (EPSRC), U.K.
Zitieren
APA
Zheng, Z., Deng, Y., Yi, W., Shin, H., & Nallanathan, A. (2025). Over-the-Air Computation Enabled Semi-Asynchronous Wireless Federated Learning. IEEE Transactions on Communications, 73(10), 8919-8936. https://doi.org/10.1109/tcomm.2025.3582727
Vancouver
Zheng Z, Deng Y, Yi W, Shin H, Nallanathan A. Over-the-Air Computation Enabled Semi-Asynchronous Wireless Federated Learning. IEEE Transactions on Communications. 2025;73(10):8919-8936. doi:10.1109/tcomm.2025.3582727
BibTeX
@article{Zheng2025,
author = {Zheng, Zijian and Deng, Yansha and Yi, Wenqiang and Shin, Hyundong and Nallanathan, Arumugam},
title = {Over-the-Air Computation Enabled Semi-Asynchronous Wireless Federated Learning},
journal = {IEEE Transactions on Communications},
year = {2025},
volume = {73},
number = {10},
pages = {8919--8936},
doi = {10.1109/tcomm.2025.3582727}
}
EUVIMED