Graph Spectral Client Scheduling for Reliable Federated Learning in Safety-Critical LEO Satellite Networks

Creative Commons License

Ahmad B.

Machine Learning, sa.2026115214, ss.1-25, 2026 (Hakemli Dergi)

Özet

Federated learning (FL) over low Earth orbit (LEO) satellite constellations faces a fundamental safety challenge. Intermittent satellite visibility causes standard client scheduling methods to systematically exclude certain geographic regions, producing ground stations with zero recall that render entire populations invisible to safety critical event detection. This work demonstrates that existing schedulers, including random selection, loss based prioritization, and operational heuristics such as maximum visibility and geographic diversity, routinely produce this catastrophic failure mode under realistic orbital dynamics, yet the failure remains hidden when only aggregate metrics are reported. Standard evaluation, relying solely on aggregate metrics such as average test accuracy and convergence speed, conceals catastrophic localized failures that disaggregated per node safety metrics reveal. To substantially reduce this failure, a graph spectral scheduling framework called Graph Centric Periodic Scheduling (GCPS) is introduced. GCPS preserves the algebraic connectivity of the training subgraph by jointly optimizing betweenness centrality, participation diversity, eigenvector contribution to the Fiedler vector, and node degree, with weights dynamically adapted through a validation only multi armed bandit exploring seven Pareto optimal weight configurations. The central finding establishes that GCPS substantially reduces zero recall nodes to 0 out of 5 trials (0%), compared to 5 out of 5 trials (100%) for geographic diversity scheduling and 2 out of 5 trials (40%) for state of the art fairness methods, namely Agnostic Federated Learning (AFL) and q Fair Federated Learning (qFFL). This substantial improvement is unmatched by any of the nine baseline methods, while GCPS also significantly reduces false negative rates (p< 0.01 versus geographic diversity, graph weighted sampling, q-FFL, AFL, and Oort), maintains competitive standard accuracy, and demonstrates robustness to topology mismatch, differential privacy noise, and label flipping attacks. Beyond the specific satellite domain, this work argues that minimum per node performance guarantees must become a primary evaluation criterion for safety critical FL, as aggregate metrics conceal catastrophic localized failures that graph spectral scheduling is uniquely positioned to prevent. It is further demonstrated that commonly used fairness indices, with Jain’s index exceeding 0.98 for all methods, measure participation equity rather than performance equity, and therefore fail to detect zero recall failures, a critical distinction for safety critical deployment.