CroSatFL: Energy-Efficient Federated Learning with Cross-Aggregation for Satellite Edge Computing

TL;DR

CroSatFL is a hierarchical federated learning framework for satellite edge computing that significantly reduces energy consumption and communication costs while maintaining high training performance in LEO satellite constellations.

Contribution

It introduces a novel on-orbit hierarchical FL framework with energy-aware mechanisms that optimize satellite cluster formation, straggler mitigation, and cross-aggregation, reducing ground station communication and energy use.

Findings

Reduces GS communication count by over 100x.

Decreases GS transmission energy by approximately 6x.

Achieves competitive accuracy and faster convergence in satellite FL.

Abstract

Low Earth Orbit (LEO) mega-constellations extend the cloud-to-edge continuum into space, enabling satellite edge computing. However, Federated Learning (FL) in this environment is fundamentally energy-constrained due to dynamic inter-satellite connectivity, heterogeneous onboard computing hardware, and strict power budgets. We propose CroSatFL, a sustainable on-orbit hierarchical FL framework that reduces end-to-end energy across computation and communication while maintaining strong training performance under realistic LEO dynamics. CroSatFL keeps the ground station (GS) off the iterative loop by performing all local training and intermediate aggregations on orbit, requiring only two GS communication phases: one for initialization and one for final model collection. This sharply reduces repeated use of bandwidth-limited and energy-expensive GS links and shifts iterative exchanges to laser inter-satellite links (LISLs). CroSatFL integrates three energy-aware mechanisms: StarMask forms LISL-feasible clusters that align data volume with heterogeneous CPU/GPU capability, Skip-One mitigates transient stragglers by skipping at most one slow client per cluster to lower round energy and latency while preserving long-term fairness, and random-k cross-aggregation enables lightweight topology-aware cross-cluster mixing without extending round duration. Using an end-to-end energy model with a realistic Walker-Delta constellation, we show that CroSatFL reduces GS communication count by over two orders of magnitude and GS transmission energy by about 6x relative to GS-centric and on-orbit baselines, while achieving competitive accuracy and faster convergence.