Over-the-Air Aggregation for Federated Learning: Waveform Superposition and Prototype Validation
Journal of Communications and Information Networks · 2021-12 · 6 (4) · 429-442
DOI: 10.23919/jcin.2021.9663107 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Communications and Information Networks, 2021-12. Verfasst von Huayan Guo und 7 weiteren Autor:innen. Die Arbeit wurde bisher 48-mal zitiert (Crossref).
Abstract (Originalsprache)
In this paper, we develop an orthogonal-frequency-division-multiplexing (OFDM)-based over-the-air (OTA) aggregation solution for wireless federated learning (FL). In particular, the local gradients in massive IoT devices are modulated by an analog waveform and are then transmitted using the same wireless resources. To this end, achieving perfect waveform superposition is the key challenge, which is difficult due to the existence of frame timing offset (TO) and carrier frequency offset (CFO). In order to address these issues, we propose a two-stage waveform pre-equalization technique with a customized multiple access protocol that can estimate and then mitigate the TO and CFO for the OTA aggregation. Based on the proposed solution, we develop a hardware transceiver and application software to train a real-world FL task, which learns a deep neural network to predict the received signal strength with global positioning system information. Experiments verify that the proposed OTA aggregation solution can achieve comparable performance to offline learning procedures with high prediction accuracy.
Zitieren
APA
Guo, H., Zhu, Y., Ma, H., Lau, V. K. N., Huang, K., Li, X., Nong, H., & Zhou, M. (2021). Over-the-Air Aggregation for Federated Learning: Waveform Superposition and Prototype Validation. Journal of Communications and Information Networks, 6(4), 429-442. https://doi.org/10.23919/jcin.2021.9663107
Vancouver
Guo H, Zhu Y, Ma H, Lau VKN, Huang K, Li X, et al. Over-the-Air Aggregation for Federated Learning: Waveform Superposition and Prototype Validation. Journal of Communications and Information Networks. 2021;6(4):429-442. doi:10.23919/jcin.2021.9663107
BibTeX
@article{Guo2021,
author = {Guo, Huayan and Zhu, Yifan and Ma, Haoyu and Lau, Vincent K. N. and Huang, Kaibin and Li, Xiaofan and Nong, Huabin and Zhou, Mingyu},
title = {Over-the-Air Aggregation for Federated Learning: Waveform Superposition and Prototype Validation},
journal = {Journal of Communications and Information Networks},
year = {2021},
volume = {6},
number = {4},
pages = {429--442},
doi = {10.23919/jcin.2021.9663107}
}
EUVIMED