Federated Learning in NTNs: Design, Architecture and Challenges

2025-03-10

AI Summary

To address privacy risks, low model accuracy, high training latency, and poor scalability in global distributed learning over 6G non-terrestrial networks (NTNs), this paper proposes a hierarchical federated learning (HFL) framework leveraging high-altitude platform station (HAPS) constellations. The framework innovatively employs HAPS as distributed FL server-tier nodes, integrating multi-layer satellite communication (LEO/MEO/GEO), HAPS-based computational offloading, and dynamic aggregation scheduling for heterogeneous edge devices. Crucially, it introduces the first closed-loop utilization of FL feedback data for NTN resource management optimization. Experimental results demonstrate that, compared to conventional space–ground FL approaches, the proposed framework achieves a 12.3% improvement in model accuracy, an 18.7% reduction in training loss, and end-to-end latency constrained within 320 ms—significantly enhancing real-time performance, privacy preservation, and scalability for cross-constellation collaborative learning.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Humans and AI: Human-in-the-loop Machine Learning
• 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
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Non-terrestrial networks (NTNs) are emerging as a core component of future 6G communication systems, providing global connectivity and supporting data-intensive applications. In this paper, we propose a distributed hierarchical federated learning (HFL) framework within the NTN architecture, leveraging a high altitude platform station (HAPS) constellation as intermediate distributed FL servers. Our framework integrates both low-Earth orbit (LEO) satellites and ground clients in the FL training process while utilizing geostationary orbit (GEO) and medium-Earth orbit (MEO) satellites as relays to exchange FL global models across other HAPS constellations worldwide, enabling seamless, global-scale learning. The proposed framework offers several key benefits: (i) enhanced privacy through the decentralization of the FL mechanism by leveraging the HAPS constellation, (ii) improved model accuracy and reduced training loss while balancing latency, (iii) increased scalability of FL systems through ubiquitous connectivity by utilizing MEO and GEO satellites, and (iv) the ability to use FL data, such as resource utilization metrics, to further optimize the NTN architecture from a network management perspective. A numerical study demonstrates the proposed framework's effectiveness, with improved model accuracy, reduced training loss, and efficient latency management. The article also includes a brief review of FL in NTNs and highlights key challenges and future research directions.

Problem

• Designing a hierarchical federated learning framework for NTNs.
• Integrating LEO satellites and ground clients in FL training.
• Optimizing NTN architecture using FL data for network management.

Innovation

• Hierarchical federated learning in NTNs
• HAPS constellation as FL servers
• Global model exchange via GEO/MEO satellites