Deep Reinforcement Learning Based Computing Offloading and Resource Allocation Algorithm for Mobile Edge Networks

Abstract

With the rapid development of Internet, continuous emergence of various innovative applications makes current mobile network face pressure of lower latency and computing capability. Mobile edge computing (MEC) has been proposed to be a promising solution to reduce the delay of interaction between applications and compensate the deficiencies of traditional cloud computing. In this paper, we propose a computing offloading and resource allocation algorithm to deal with problems in mobile edge networks (MEN), including offloading decision, transmission power and computation resources allocation. With the goal of minimizing the total cost of the system, an algorithm combining Deep Reinforcement Learning (DRL) and Genetic Algorithm (GA) is used to obtain an approximate optimal solution for the system. Simulation results prove the effectiveness of the algorithm.