Energy-efficient Resource Allocation for UAV-empowered Mobile Edge Computing System

Unmanned aerial vehicles (UAVs) have been gained significant attention from mobile network operators (MNOs) to provision low-latency wireless big data applications, where a number of ground resource-limited user equipments (UEs) can be served by UAVs equipped with powerful computing resources, in comparison with UEs. In this paper, a novel UAV-empowered mobile edge computing (MEC) network architecture is considered. An energy consumption and task execution delay minimization multi-objective optimization problem is formulated, subject to numerous QoS constraints. A heuristic algorithm is proposed to solve the challenging optimization problem, which consists of the task assignment, differential evolution (DE)-aided and non-dominated sort steps. The selected key performance of the proposed algorithm is given and compared with the existing advanced particle swarm optimization (PSO) and non-dominated sorting genetic algorithm II (NSGA-II). The results show that the proposed heuristic algorithm promises higher energy efficiency than PSO and NSGA-II under the same task execution time cost.