Graph-guided MADQN strategy improves handover in LEO satellite networks

This article addresses the challenge of frequent communication link handovers in low earth orbit (LEO) satellite networks caused by high satellite velocity. It proposes a graph-guided MADQN-based handover strategy to improve communication continuity by balancing handover frequency, signal quality, and network load. The approach aims to overcome limitations of traditional fixed-rule and reinforcement learning methods, such as low training efficiency and unstable convergence in high-dimensional state spaces.

Originally published by gnews