Ground-Assisted Federated Learning in LEO Satellite Constellations

TL;DR

This paper introduces a novel asynchronous federated learning algorithm tailored for LEO satellite constellations, demonstrating improved robustness and fast convergence in image classification tasks.

Contribution

It proposes a new asynchronous FL algorithm based on FedAvg specifically designed for LEO satellite networks, enhancing robustness and convergence speed.

Findings

Fast convergence speed demonstrated on MNIST and CIFAR-10 datasets

Improved robustness against heterogeneous satellite scenarios

High asymptotic test accuracy achieved

Abstract

In Low Earth Orbit (LEO) mega constellations, there are relevant use cases, such as inference based on satellite imaging, in which a large number of satellites collaboratively train a machine learning model without sharing their local datasets. To address this problem, we propose a new set of algorithms based on Federated learning (FL), including a novel asynchronous FL procedure based on FedAvg that exhibits better robustness against heterogeneous scenarios than the state-of-the-art. Extensive numerical evaluations based on MNIST and CIFAR-10 datasets highlight the fast convergence speed and excellent asymptotic test accuracy of the proposed method.