An Edge Cloud Resource Scheduling Mechanism in the Space-Air-Terrestrial-Sea Integrated Network

Edge computing enables services and functions that would otherwise be located in the cloud to be moved near the users, providing powerful computing, storage, networking and communication capabilities at the edge of the network. However, edge computing is usually deployed on the ground or in Unmanned Aerial Vehicle (UAV) networks, and how the communication range scales is a key issue. Space-Air-Terrestrial-Sea Integrated Network (SATSIN) aims to provide wide-area connection and efficient processing capability for 5G/6G communication. Among them, satellite network, as the key link to realize global communication connection, has attracted wide attention. In this paper, we study the resource scheduling problem in the context of SATSIN. The goal is to achieve approximate optimal performance and reduce computational delay while considering the constraints of actual satellite communication and satellite force resources. We propose an edge cloud model based on SATSIN environment, in which satellites equipped with computing resources can provide offloading services to users in the communication window. Considering the non-convexity and state space of on-satellite edge cloud resource scheduling problem, we propose an edge computing offloading strategy based on Deep Reinforcement Learning (DRL). Experimental results show that this scheme can improve the system performance and realize dynamic update and offloading decision.