Edge Server Deployment Based on Reinforcement Learning in Mobile Edge Computing

Mobile Edge Computing (MEC) is to sink the resource of remote cloud computing center to the edge network to provide users with better services. Distinct from cloud computing, mobile edge computing is under the constraint of edge server computing resources, deployment location, wireless transmission bandwidth and etc. Although there has been significant research in the field of mobile edge computing, little attention has been given to understanding the placement of edge servers to optimize the mobile edge computing network performance. In this paper, we propose a server deployment scheme based on reinforcement learning. Firstly, we propose the MEC three-tier architecture, which takes the latency of the base station as the optimization goal. Then, we formulate the edge server deployment problem as a single-objective optimization problem, and propose the edge server deployment (Q-ESD) algorithm in this paper based on the Q-Learning algorithm. Finally, experimental results show that our approach outperforms several representative approaches in terms of access delay and workload balancing.