DDPG Based Task Offloading Scheme for Optimal Unmanned Aerial Vehicle Security Communication Benefit
Due to the advantages of high mobility and flexibility,Unmanned Aerial Vehicles(UAVs)are widely used in mobile edge computing systems.However,its line of sight air to ground wireless link makes drones more susceptible to enemy eavesdropping.To solve this problem,a multi-user and multi task UAV assisted Mobile Edge Computing(MEC)security communication system model consisting of auxiliary UAV,user equipment,intruder UAV and base station is constructed firstly.Then,by introducing the concept of security communication benefit,the problem of UAV privacy security communication is transformed into the problem of maximizing the security communication benefit of the auxiliary UAV.Finally,a task offloading algorithm based on Deep Deterministic Policy Gradient(DDPG)for optimal security communication benefit is proposed.This algorithm maximizes the security communication benefit of drones by jointly optimizing the release position of drones,user task offloading rate,and allocation of computing resources between base stations and drones.The simulation results show that the security communication benefits achieved by this algorithm are 25.42％higher on average compared to other algorithms in terms of different task volumes,and 33.12％higher on average in terms of different task complexity.
security communication benefitsDDPGMECUAVresource allocation
国家科技期刊平台
NSTL
万方数据
