Computing Offloading Based on Path Planning in Satellite Terrestrial Integrated Network

The satellite-terrestrial integrated network (STIN) integrated with mobile edge computing (MEC) can effectively improve the overall utilization rate of network resources and provide users with all-round computing services and diversified access services due to its advantages of wide coverage, large capacity and flexible networking. However, because of problems such as the dynamic topology of the STIN, the diversity of user service requirements, and the high deployment cost of edge servers, the network quality-of-service (QoS) requirements of users for computation-intensive and delay-sensitive services cannot be met. In order to solve the problem mentioned above, this paper proposes a task offloading based on path planning (TOBPP) scheme. First, considering the collaborative processing of satellite-terrestrial edge nodes, a double mobile edge computing (DMEC) framework is proposed. Next, a task processing delay model is constructed, and the problem of minimizing the task processing delay under the constraints of storage resources, offloading decisions and mapping conditions is formed. Then, it is decoupled into two sub-problems: minimizing the task-completion delay of terrestrial users and minimizing the task-completion delay of satellite users. Finally, a hybrid distributed (HD) algorithm is proposed to solve the problem. Simulation results show that the proposed TOBPP scheme can improve the Quality of Experience (QoE) for real-time applications while reducing the deployment cost of edge servers.