FedSpace: An Efficient Federated Learning Framework at Satellites and Ground Stations

TL;DR

FedSpace is a novel federated learning framework designed for satellite-ground station networks, effectively addressing connectivity challenges and reducing training time for Earth imagery analysis.

Contribution

The paper introduces FedSpace, a new FL framework that dynamically schedules model aggregation considering satellite orbits and connectivity, improving training efficiency.

Findings

FedSpace reduces training time by 38.6% compared to existing FL methods.

It effectively manages satellite-ground connectivity variability in FL.

Numerical evaluations demonstrate significant efficiency gains.

Abstract

Large-scale deployments of low Earth orbit (LEO) satellites collect massive amount of Earth imageries and sensor data, which can empower machine learning (ML) to address global challenges such as real-time disaster navigation and mitigation. However, it is often infeasible to download all the high-resolution images and train these ML models on the ground because of limited downlink bandwidth, sparse connectivity, and regularization constraints on the imagery resolution. To address these challenges, we leverage Federated Learning (FL), where ground stations and satellites collaboratively train a global ML model without sharing the captured images on the satellites. We show fundamental challenges in applying existing FL algorithms among satellites and ground stations, and we formulate an optimization problem which captures a unique trade-off between staleness and idleness. We propose a novel FL framework, named FedSpace, which dynamically schedules model aggregation based on the deterministic and time-varying connectivity according to satellite orbits. Extensive numerical evaluations based on real-world satellite images and satellite networks show that FedSpace reduces the training time by 1.7 days (38.6%) over the state-of-the-art FL algorithms.