ACO-based Optimal Node Selection Method for QoE Improvement in MEC Environment
Sanghoon Lee, Hwa‐Sung Kim · 2019 International Conference on Electronics, Information, and Communication (ICEIC) · 2019
Mobile Edge Computing (MEC) is a technology that provides cloud computing services with ultra-low latency and large bandwidth by placing distributed cloud computing server near each wireless base stations so that various services and caching contents are deployed close to user mobile devices. However, if a network failure occurs at a point in a wireless base station or many mobile devices are crowded at a point, the task failure rate of applications that offloaded the tasks here is increased. Therefore, users experience a low quality of experience (QoE). In this paper, we propose an optimal node selection method based on ACO (Ant Colony Optimization) to solve this problem. In the MEC environment, when mobile devices are crowded around a certain base station or when network congestion occurs, tasks are autonomously offloaded to the optimal node among the nearby nodes monitored in real time through the ACO. This provides users with low task failure rate and low latency delay despite of a lack of resources or of network congestion in a specific base station. Therefore, it provides improved QoE by lowering network delay time and task failure rate. In order to prove this, performance analysis was performed by using EdgeCloudsim. We can confirm that the proposed ACO based optimal node selection algorithm obtained better performance through lower task failure rate and delay time compared to other scenarios.