Cooperative Caching and Beamforming for CoMP-Enabled Mobile Edge Networks

Abstract

Mobile edge caching and coordinated multi-point (CoMP) joint transmission (JT) techniques are expected to further reduce content delivery delay. In this paper, we study the joint optimization of content caching, user association and beamforming coordination in CoMP-enabled mobile edge networks, aiming to minimize the weighted sum of content delivery delay. Based on double time-scale scheduling model, we formulate the joint optimization problem as a Markov decision process (MDP), and develop a deep reinforcement learning (DRL) based joint cooperative caching and beamforming coordination (JCCBC) algorithm. With the double time-scale scheduling method, decisions are made at different time-scale respectively. In the proposed JCCBC algorithm, decisions on user association and beamforming coordination are optimized in small time scale based on the time-varying available radio resources of different base stations (BSs), so as to minimize the transmission delay. Meanwhile, cooperative caching decisions are optimized in large time scale to reduce the retrieval delay according to user requests and wireless communication environment. Simulation results demonstrate that the proposed JCCBC algorithm can effectively reduce the content delivery delay, as compared to benchmark schemes.