Joint Computing Resource Scheduling and Task Priority Selection in UAV-Enabled MEC

Owing to the outstanding characteristics of unmanned aerial vehicles (UAV), i.e., mobility and flexibility, UAV-enabled mobile edge computing (MEC) has become a widely attractive research direction. This paper studies the computing offload problem of multi UAV auxiliary terminal devices (TDs), in which multiple UAVs equipped with computing resources help the terminal device with limited local computing resources and energy to accomplish computing tasks. There are three computational strategies for the task of terminal devices. Firstly, each terminal device computes the task by itself. Secondly, each task can be offloaded to an UAV for computation. Thirdly, each task can be offloaded to nearby base stations (BS) for computation through the UAV relay. In this paper, we propose a partial offloading method and determine the task segmentation threshold. Due to the difference in computing power between heterogeneous UAVs and terminal devices, different tasks have different requirements for the time delay. The research problem is a large-scale nonlinear and complex problem, which is challenging to solve. Therefore, the terminal device is divided into different regions. Then we assign the UAV with strong computing power to the regions with weak computing power to achieve the complementarity of computing power. In addition, we prioritize the tasks of the end devices. Based on this, we propose the MST algorithm. Simulation results prove that our scheme significantly outperforms the baseline approach in terms of task completion time and compared to the SCA method, Our proposed approach improves performance by 13% on large-scale tasks.