Priority-Aware Task Offloading and Resource Allocation in Satellite and HAP Assisted Edge-Cloud Collaborative Networks

Mobile edge computing (MEC) has been regarded as an effective technique to handle computing intensive tasks of IoT devices (IoTDs). In this paper, we present a priority-aware edge computing model for space-air-ground integrated networks (SAGIN), which contain a satellite, a high altitude platform (HAP) and numerous IoTDs. In such networks, the HAP provides edge computing services close to the IoTDs and the cloud servers connected to satellite provide cloud computing services. The network considers different services for IoTDs with various Quality-of-Service (QoS) requirements. First, we design a media access control MAC protocol named priority-aware time division multiple access (PTDMA) for controlling the large-scale access of IoTDs. Second, for tasks with different priorities, we propose a priority-aware deep deterministic policy gradient (DDPG) based task offloading and resource allocation scheme (PDOA). Simulation experimental results validate the performance of the PTDMA and PDOA scheme, which can effectively improve the average system utility in a stochastic time-varying environment.