Federated Learning-driven Beam Management in LEO 6G Non-Terrestrial Networks

2026-03-11

AI Summary

This work addresses the challenge of inefficient beam management in low Earth orbit (LEO) non-terrestrial networks operating under dynamic propagation conditions. To overcome this limitation, the authors propose an intelligent beam selection mechanism based on federated learning, leveraging high-altitude balloon platforms to enable distributed collaborative training within orbital planes. The study innovatively integrates graph neural networks (GNNs) into federated beam management for LEO satellites, jointly optimizing beam selection using real-world channel and beamforming data. Experimental results demonstrate that the proposed GNN-based model significantly outperforms conventional multilayer perceptrons (MLPs) in both beam prediction accuracy and stability, with particularly pronounced performance gains observed in low-elevation-angle scenarios.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for 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

Low Earth Orbit (LEO) Non-Terrestrial Networks (NTNs) require efficient beam management under dynamic propagation conditions. This work investigates Federated Learning (FL)-based beam selection in LEO satellite constellations, where orbital planes operate as distributed learners through the utilization of High-Altitude Platform Stations (HAPS). Two models, a Multi-Layer Perceptron (MLP) and a Graph Neural Network (GNN), are evaluated using realistic channel and beamforming data. Results demonstrate that GNN surpasses MLP in beam prediction accuracy and stability, particularly at low elevation angles, enabling lightweight and intelligent beam management for future NTN deployments.

Problem

• Beam Management
• Low Earth Orbit
• Non-Terrestrial Networks
• Dynamic Propagation
• 6G

Innovation

• Federated Learning
• Beam Management
• LEO NTN
• Graph Neural Network
• HAPS