Quality of Service Optimization in Mobile Edge Computing Networks via Deep Reinforcement Learning
Li-Tse Hsieh, Hang Liu, Yang Guo, Robert Gazda
Mobile edge computing (MEC) is an emerging paradigm that integrates computing resources in wireless access networks to process computational tasks in close proximity to mobile users with low latency. In this paper, we propose an online double deep Q networks (DDQN) based learning scheme for task assignment in dynamic MEC networks, which enables multiple distributed edge nodes and a cloud data center to jointly process user tasks to achieve optimization of the expected long-term quality of service (QoS). The proposed scheme can capture a wide range of dynamic network parameters including non-stationary task arrivals, node computing capabilities, and network delay statistics, and learn the optimal task assignment policy with no assumption on the knowledge of the underlying dynamics. The proposed algorithm accounts for both performance and complexity, and ad-dresses the state and action space explosion problem in conventional Q learning. The evaluation results show that the proposed DDQN-based task assignment scheme significantly improves the QoS performance, compared to the existing schemes that do not consider the effects of network dynamics on the expected long-term rewards and can scale reasonably well as the network size increases.
September 13-15, 2020
The 15th International Conference on Wireless Algorithms, Systems, and Applications (WASA 2020)