Multi-Channel Two-Step Matching in High-Density LEO Satellites Network for Diverse User Demands

Abstract

This letter addresses the wireless resource scheduling challenge for high-density low-earth orbit (LEO) satellite constellations, catering to the demands of wide-area User Equipments (UEs). Aiming to maximize system capacity, we conceptualize the issue as a multi-branch bipartite graph matching problem with specially designed vertices, and propose an ‘access-right’ based two-step matching mechanism for satellite channel allocation (ARSMM). ARSMM runs in two steps. Firstly, the mechanism employs a Hungarian algorithm based on determining ‘access-right’ to obtain satellite-UE scheduling. Secondly, it introduces a matching adjustment algorithm based on three criteria, respectively, to accommodate the varying channel demands of the UEs. Numerical simulation results indicate that ARSMM markedly improves resource utilization efficiency and system capacity compared to the baseline algorithm. Moreover, it achieves near-parity with the Genetic Algorithm while significantly improving processing speed to millisecond-level efficiency.