Communication-Efficient Satellite-Ground Federated Learning Through Progressive Weight Quantization
IEEE Transactions on Mobile Computing · 2024-09 · 23 (9) · 8999-9011
DOI: 10.1109/tmc.2024.3358804 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2024-09. Verfasst von Chen Yang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 52-mal zitiert (Crossref) und nennt 60 Literaturangaben. Gefördert von: National Key R&D Program of China, Shenzhen Science and Technology Innovation Commission Free Exploring Basic Research Project, National Natural Science Foundation of China.
Abstract (Originalsprache)
Large constellations of Low Earth Orbit (LEO) satellites have been launched for Earth observation and satellite-ground communication, which collect massive imagery and sensor data. These data can enhance the AI capabilities of satellites to address global challenges such as real-time disaster navigation and mitigation. Prior studies proposed leveraging federated learning (FL) across satellite-ground to collaboratively train a share machine learning (ML) model in a privacy-preserving mechanism. However, they mostly focus on single unique challenges such as limited ground-to-satellite bandwidth, short connection window, and long connection cycle, while ignoring the completeness of these challenges in deploying efficient FL frameworks in space. In this paper, we propose an efficient satellite-ground FL framework, SatelliteFL, to address these three challenges collectively. Its key idea is to ensure that each satellite must complete per-round training within each connection window. Moreover, we design a progressive block-wise quantization algorithm that determines a unique bitwidth for each block of the ML model to maximize the model utility while not exceeding the connection window. We evaluate SatelliteFL by plugging an implemented FL platform into real-world satellite networks and satellite images. The results show that SatelliteFL highly accelerates the convergence by up to 2.8× and improves the bandwidth utilization ratio by up to 9.3× compared to the state-of-the-art methods.
Förderung
- National Key R&D Program of China
- Shenzhen Science and Technology Innovation Commission Free Exploring Basic Research Project
- National Natural Science Foundation of China
Zitieren
APA
Yang, C., Yuan, J., Wu, Y., Sun, Q., Zhou, A., Wang, S., & Xu, M. (2024). Communication-Efficient Satellite-Ground Federated Learning Through Progressive Weight Quantization. IEEE Transactions on Mobile Computing, 23(9), 8999-9011. https://doi.org/10.1109/tmc.2024.3358804
Vancouver
Yang C, Yuan J, Wu Y, Sun Q, Zhou A, Wang S, et al. Communication-Efficient Satellite-Ground Federated Learning Through Progressive Weight Quantization. IEEE Transactions on Mobile Computing. 2024;23(9):8999-9011. doi:10.1109/tmc.2024.3358804
BibTeX
@article{Yang2024,
author = {Yang, Chen and Yuan, Jinliang and Wu, Yaozong and Sun, Qibo and Zhou, Ao and Wang, Shangguang and Xu, Mengwei},
title = {Communication-Efficient Satellite-Ground Federated Learning Through Progressive Weight Quantization},
journal = {IEEE Transactions on Mobile Computing},
year = {2024},
volume = {23},
number = {9},
pages = {8999--9011},
doi = {10.1109/tmc.2024.3358804}
}
EUVIMED