Cooperative Computation Offloading for UAVs: A Joint Radio and Computing Resource Allocation Approach

Research and applications of unmanned aerial vehicles (UAVs) are becoming increasingly prosperous in these years due to the maturity of the aircraft technology and regulations. A large amount of UAVs are to be deployed in cities to undertake tasks such as environment monitoring and security surveillance. For those computation-intensive tasks, on-board execution can lead to inefficiency and unsustainability due to the limited battery life and computing resources of UAVs. To this end, this paper adopts cooperative mobile edge computing such that energy consumption and task execution latency can both be reduced. The computation offloading for UAVs aims to optimize the energy and latency jointly with the help of cooperative edge servers. We obtain the most energy efficient offloading data rate by convex optimization and obtain the optimal data allocation scheme to meet the latency constraint by simulated annealing based particle swarm optimization (SAPSO). Simulation results validate the efficiency of the proposed UAV computation offloading strategy.