Efficient Resource Management Based on DQN in LEO Satellite Edge Computing System

In this paper, a low earth orbit (LEO) satellite edge computing architecture is proposed by placing the mobile edge computing (MEC) servers on the LEO satellites to provide computing services for ground users. To solve the offloading decision and resource allocation problems in multi-user, multi-LEO satellite and multi-task scenario, we propose a computation offloading algorithm based on deep reinforcement learning (DRL), derive the sub-optimal power allocation scheme, and obtain the optimal MEC resource allocation scheme through convex optimization algorithm. Simulation results show that the proposed algorithm achieves excellent convergence effect and system performance.