SatFed: A Resource-Efficient LEO Satellite-Assisted Heterogeneous Federated Learning Framework

TL;DR

SatFed introduces a resource-efficient satellite-assisted federated learning framework that leverages LEO satellite networks to overcome terrestrial network limitations, optimizing model updates and handling device heterogeneity effectively.

Contribution

The paper proposes SatFed, a novel framework that uses model prioritization and a multigraph approach to improve federated learning over satellite networks with heterogeneous devices.

Findings

SatFed outperforms existing benchmarks in accuracy and robustness.

It effectively manages limited satellite-ground bandwidth.

The framework enhances local training through satellite-assisted peer guidance.

Abstract

Traditional federated learning (FL) frameworks rely heavily on terrestrial networks, where coverage limitations and increasing bandwidth congestion significantly hinder model convergence. Fortunately, the advancement of low-Earth orbit (LEO) satellite networks offers promising new communication avenues to augment traditional terrestrial FL. Despite this potential, the limited satellite-ground communication bandwidth and the heterogeneous operating environments of ground devices-including variations in data, bandwidth, and computing power-pose substantial challenges for effective and robust satellite-assisted FL. To address these challenges, we propose SatFed, a resource-efficient satellite-assisted heterogeneous FL framework. SatFed implements freshness-based model prioritization queues to optimize the use of highly constrained satellite-ground bandwidth, ensuring the transmission of the most critical models. Additionally, a multigraph is constructed to capture real-time heterogeneous relationships between devices, including data distribution, terrestrial bandwidth, and computing capability. This multigraph enables SatFed to aggregate satellite-transmitted models into peer guidance, enhancing local training in heterogeneous environments. Extensive experiments with real-world LEO satellite networks demonstrate that SatFed achieves superior performance and robustness compared to state-of-the-art benchmarks.