Task Delay Minimization in Wireless Powered Mobile Edge Computing Networks: A Deep Reinforcement Learning Approach

Abstract

With the emergence of a large number of intelligent terminals, the network is facing severe challenges such as insufficient resources, limited bandwidth and environmental pollution. This paper combines mobile edge computing (MEC) and wireless power transfer (WPT) technologies. Firstly, in order to minimize task delay, a wireless powered mobile edge computing network (WPMECN) model based on binary offloading strategy is constructed. Then, a deep reinforcement learning (DRL) framework is designed to minimize the network task delay. Meanwhile, wireless resource allocation and task offloading decision are jointly optimized. Finally, simulation experiments are carried out with Pytorch tool. It is verified that the proposed scheme is able to guarantee similar delay performance achieved by global searching scheme, while the scheme operation delay of the proposed scheme is much lower.