A Joint Resource Allocation and Task Offloading Algorithm in Satellite Edge Computing

This paper studies the task offloading problem for ground users in remote areas in satellite edge computing. Each user can offload computation tasks to either the Geosynchronous Earth Orbit (GEO) satellite, forward them to the ground cloud computing center, or offload them to a Low Earth Orbit (LEO) satellite which is constantly moving relative to the ground. To obtain the optimal task offloading plan and resource allocation plan that minimize system computing delay, we formulate this problem as a mixed integer nonlinear programming (MINLP) problem and propose a low complexity solution algorithm for it. Through mathematical derivation, we can organize the MINLP problem into three separate solutions: optimal allocation of computing resources, optimal transmission power control, and optimal offloading plan. In our algorithm, we apply the Lagrange multiplier method and binary search to obtain the optimal allocation of computing resources and optimal transmission power control under a given offloading plan. Then, using our proposed method based on the idea of greedy algorithm, we obtain an approximate optimal solution for task offloading. Compared to other algorithms, our proposed algorithm significantly reduces the system cost with a low computation complexity.