A Cooperative Community-Based Framework for Service Caching and Task Offloading in Multi-Access Edge Computing

Zhuofan Liao,GuiYing Yin,XiaoYong Tang,Penglu Liu

DOI: https://doi.org/10.1109/tnsm.2024.3372295

2024-01-01

IEEE Transactions on Network and Service Management

Abstract:In multi-access edge computing, services are cached from cloud servers to the edge base station providing low-latency services. Horizontal Collaboration (HC) between Base Stations (BSs) can alleviate the problem of resource limitation in edge base stations. However, the initiative of cooperation between base stations and the interests of base stations themselves are not guaranteed. In this paper, we study the Joint Service Caching and Task Offloading decision Problem (JSCTOP). To solve this problem, we propose a Two-Stage Optimization Framework (TSOF) based on cooperative community idea to maximize the benefit of each base station. In the first stage, a Collaborative Community Mechanism (CCM) is designed based on the Dissimilarity of Service Types (DST) to enhance the collaborative capability of BSs. In the second stage, an algorithm for service caching and task decision to update and replace the Service Types (STs) of BSs. We also design an algorithm for service pricing to incentivize BSs. In this algorithm, BSs are able to profit from providing services to users and also profit from assisting other base stations. Extensive simulation results show that TSOF outperforms its counterparts in average delay, total energy, total benefit, and community.

computer science, information systems