Task Allocation Optimization Strategy in UAV-enabled Mobile Edge Computing System

Unmanned Aerial Vehicles (UAVs) have been widely used as aerial computing devices in Internet of Things (IoT) to support time-sensitive and compute-intensive applications and services, e.g. automatic driving, data analysis, etc. In this paper, a UAV-enabled mobile edge computing (MEC) system is considered for computing offloading services, in which the offloaded tasks from ground devices are processed by UAVs. However, when multiple computing tasks exist, the task allocation for devices should be carried out at the lowest computing cost. To this end, we first propose a non-cooperative game model-based power allocation (NGPA) scheme, and the optimal power allocation strategy of devices is determined to minimize transmission energy consumption. Next, we model the problem of task allocation as a bilateral matching between devices and UAVs to minimize the overall computing cost. Then, a bilateral matching mechanism based on improved Hungarian algorithm is provided to obtain the optimal solution. Finally, the simulation results show that the proposed scheme outperforms the benchmark schemes on computing cost.