Decentralized computation offloading for multi-user mobile edge computing: a deep reinforcement learning approach

• Published in: EURASIP journal on wireless communications and networking Vol. 2020; no. 1; pp. 1 - 21
• Main Authors: Chen, Zhao, Wang, Xiaodong
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 29.09.2020; SpringerOpen
• Subjects: Communications Engineering; Computation offloading; Deep reinforcement learning; Engineering; Information Systems Applications (incl.Internet); Machine Learning and Deep Learning Methods for Wireless Network Applications; Mobile edge computing; Multi-user MIMO; Networks; Signal,Image and Speech Processing; Deep reinforcement learning; Computation offloading; Multi-user MIMO; Mobile edge computing
• DOI: 10.1186/s13638-020-01801-6
• ISSN: 1687-1499

Mobile edge computing (MEC) emerges recently as a promising solution to relieve resource-limited mobile devices from computation-intensive tasks, which enables devices to offload workloads to nearby MEC servers and improve the quality of computation experience. In this paper, an MEC enabled multi-user multi-input multi-output (MIMO) system with stochastic wireless channels and task arrivals is considered. In order to minimize long-term average computation cost in terms of power consumption and buffering delay at each user, a deep reinforcement learning (DRL)-based dynamic computation offloading strategy is investigated to build a scalable system with limited feedback. Specifically, a continuous action space-based DRL approach named deep deterministic policy gradient (DDPG) is adopted to learn decentralized computation offloading policies at all users respectively, where local execution and task offloading powers will be adaptively allocated according to each user’s local observation. Numerical results demonstrate that the proposed DDPG-based strategy can help each user learn an efficient dynamic offloading policy and also verify the superiority of its continuous power allocation capability to policies learned by conventional discrete action space-based reinforcement learning approaches like deep Q-network (DQN) as well as some other greedy strategies with reduced computation cost. Besides, power-delay tradeoff for computation offloading is also analyzed for both the DDPG-based and DQN-based strategies.