FedRings: A Scalable and Topology-Aware Federated Learning Framework for LEO Satellite Constellations

2026-08-04

AI Summary

This work addresses the challenges posed by frequent link handovers, brief contact windows, and highly dynamic topologies in low Earth orbit (LEO) satellite networks, which severely hinder the efficiency and scalability of federated learning. To overcome these limitations, the paper proposes FedRings, a ring-based decentralized federated learning framework tailored for LEO satellites. FedRings innovatively integrates spatiotemporal routing, link-aware scheduling, adaptive sparse incremental aggregation, and a historical compensation mechanism to efficiently align model exchanges within visibility windows and propagate updates along the ring topology. Experimental results demonstrate that FedRings significantly outperforms existing approaches in realistic LEO scenarios, achieving substantially improved training stability and scalability while reducing communication overhead.

Abstract

Federated learning over low Earth orbit (LEO) satellite networks is limited by frequent link changes, short contact times, and a highly dynamic topology, making centralized or synchronized training inefficient and hard to scale. To address this, we propose FedRings, a decentralized framework that organizes satellites into ring-based communication structures. It uses a spatio-temporal routing strategy with link-aware communication scheduling to align model exchange with actual visibility windows and time-varying connectivity patterns in LEO. Model updates are propagated along the ring using adaptive sparse incremental aggregation, which reduces communication overhead by progressively combining and compressing updates. To handle communication interruptions, a historical compensation mechanism maintains training continuity. By combining topology-aware routing, communication scheduling, and efficient aggregation, FedRings enables stable and efficient learning in dynamic LEO networks while reducing communication cost, and experiments show it consistently outperforms existing methods in realistic settings.

Problem

• federated learning
• LEO satellite constellations
• dynamic topology
• communication constraints
• scalability

Innovation

• ring-based topology
• spatio-temporal routing
• adaptive sparse aggregation
• link-aware scheduling
• historical compensation