Affiliations: [a] School of Electrical Engineering, Zhengzhou University of Science and Technology, Zhengzhou 450000, China | [b] School of Automotive Studies, Henan College of Transportation, Zhengzhou 450000, China
Abstract: Edge computing can place computing, storage, bandwidth, application and other resources on the edge of the network to reduce transmission delay, and bandwidth limitation for users. Therefore, task scheduling and resource allocation in edge computing have become a new research focus. In view of the dynamic nature of computing resources in edge computing and the computing power limitation of a single computing resource, this paper proposes a computing resource allocation scheme based on edge computing under augmented reality (AR) application. On the premise of ensuring the completion time of sub-tasks and the revenue of computing service devices, the method transforms the computing resource allocation problem in edge computing into the many-to-many matching problem between sub-tasks and computing service devices. The performance of the proposed scheme is analyzed through experiments, and the results show that the resource allocation time is relatively stable, and it is better than other state-of-the-art algorithms in terms of the edge calculation, task delay, and task violation rate.
Keywords: Edge computing, resource allocation, AR, matching problem
