A Semi-Supervised Federated Learning Framework with Hierarchical Clustering Aggregation for Heterogeneous Satellite Networks

2025-07-29

AI Summary

To address unreliable convergence, high communication overhead, and excessive energy consumption in federated learning (FL) over low-Earth-orbit (LEO) satellite networks—exacerbated by severe device heterogeneity and scarce labeled data—this paper proposes a space–ground collaborative semi-supervised hierarchical FL framework. The framework introduces a novel two-tier clustering aggregation mechanism: ground stations coordinate global model updates, while core satellites guide unlabeled edge nodes via consistency regularization and pseudo-label generation to enable semi-supervised collaboration. Additionally, parameter sparsification and adaptive weight quantization are integrated to reduce communication load. Experiments demonstrate that, while preserving model accuracy, the proposed method reduces processing time by up to 3× and system energy consumption by up to 4× compared to state-of-the-art approaches, significantly enhancing the efficiency and scalability of distributed intelligent training in LEO scenarios.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Low Earth Orbit (LEO) satellites are emerging as key components of 6G networks, with many already deployed to support large-scale Earth observation and sensing related tasks. Federated Learning (FL) presents a promising paradigm for enabling distributed intelligence in these resource-constrained and dynamic environments. However, achieving reliable convergence, while minimizing both processing time and energy consumption, remains a substantial challenge, particularly in heterogeneous and partially unlabeled satellite networks. To address this challenge, we propose a novel semi-supervised federated learning framework tailored for LEO satellite networks with hierarchical clustering aggregation. To further reduce communication overhead, we integrate sparsification and adaptive weight quantization techniques. In addition, we divide the FL clustering into two stages: satellite cluster aggregation stage and Ground Stations (GSs) aggregation stage. The supervised learning at GSs guides selected Parameter Server (PS) satellites, which in turn support fully unlabeled satellites during the federated training process. Extensive experiments conducted on a satellite network testbed demonstrate that our proposal can significantly reduce processing time (up to 3x) and energy consumption (up to 4x) compared to other comparative methods while maintaining model accuracy.

Problem

• Achieving reliable FL convergence in heterogeneous satellite networks
• Minimizing processing time and energy consumption in FL
• Handling partially unlabeled data in dynamic LEO environments

Innovation

• Semi-supervised federated learning for LEO satellites
• Hierarchical clustering aggregation in two stages
• Sparsification and adaptive weight quantization techniques