FedHC: A Hierarchical Clustered Federated Learning Framework for Satellite Networks

TL;DR

FedHC is a hierarchical federated learning framework designed for satellite networks that improves convergence speed and reduces energy consumption through satellite clustering and meta-learning-driven re-clustering.

Contribution

This paper introduces a novel hierarchical clustered federated learning framework with satellite clustering and adaptive re-clustering for satellite networks.

Findings

FedHC reduces processing time by up to 3x.

FedHC cuts energy consumption by up to 2x.

Maintains model accuracy while improving efficiency.

Abstract

With the proliferation of data-driven services, the volume of data that needs to be processed by satellite networks has significantly increased. Federated learning (FL) is well-suited for big data processing in distributed, resource-constrained satellite environments. However, ensuring its convergence performance while minimizing processing time and energy consumption remains a challenge. To this end, we propose a hierarchical clustered federated learning framework, FedHC. This framework employs a satellite-clustered parameter server (PS) selection algorithm at the cluster aggregation stage, grouping nearby satellites into distinct clusters and designating a cluster center as the PS to accelerate model aggregation. Several communicable cluster PS satellites are then selected through ground stations to aggregate global parameters, facilitating the FL process. Moreover, a meta-learning-driven satellite re-clustering algorithm is introduced to enhance adaptability to dynamic satellite cluster changes. The extensive experiments on satellite networks testbed demonstrate that FedHC can significantly reduce processing time (up to 3x) and energy consumption (up to 2x) compared to other comparative methods while maintaining model accuracy.