A Game Theory-Based Distributed Downlink Spectrum Sharing Method in Large-Scale Hybrid Satellite Constellations

Large-scale satellite constellations lead to a scarcity of spatial spectrum resources, especially for the overlapped spectrum between Low Earth Orbit ( LEO ) and Geostationary Orbit ( GEO )satellites in hybrid constellations. Hence, based on game theory, a distributed spectrum-sharing method is proposed for downlink spectrum sharing in large-scale hybrid satellite constellations. Specifically, the system cycle is divided into equal-spaced topological periods, and the beams of LEO satellites are assigned to LEO ground stations ( LGS ) during every topological period. A system model based on game theory is also developed to describe the mutual interference of links established between LEO satellites and LGS . Subsequently, the formulated game is proven to be an exact potential game, with at least one pure strategic Nash equilibrium (NE). Along the line, to obtain the NE solution, a dynamic channel allocation algorithm is proposed based on stochastic learning theory, and the convergence is proven. Finally, the simulation results demonstrate the proposed DCASLA's effectiveness, which can balance the convergence speed and overall network satisfaction.