Matching Game based Task Offloading and Resource Allocation Algorithm for Satellite Edge Computing Networks

Abstract

Aiming at the problem of computing offloading with resource constraints in edge computing of dual-layer satellite networks, each low earth orbit (LEO) satellite can offload computing workloads to Geostationary Orbit (GEO) satellites. In order to obtain the optimal strategy for radio resource allocation and offloading decision-making, the system overhead minimization problem is studied, and the offloading decision problem is modeled as a two-sided matching game problem with LEO satellites and GEO satellites as participants. The solution of the offloading decision is given by an improved two-sided many-to-one matching game algorithm with a preference for delay and energy cost. Simulation results show that by effectively offloading decision-making and power, bandwidth, and computing resource allocation, the proposed algorithm can achieve remarkable performance in satellite network edge computing. Meanwhile, compared with the existing methods, the proposed offloading algorithm can effectively reduce the running time of the offloading decision algorithm by 55.53% while the system overhead is reduced by 8.8%.