Blockchain-based Trustworthy Service Caching and Task Offloading for Intelligent Edge Computing
Yutong Zhou, Xi Li, Hong Ji, Heli Zhang · 2021 IEEE Global Communications Conference (GLOBECOM) · 2021
The upcoming 6G era involves an increasing level of data processing and capacity, where mobile edge computing (MEC) system is deployed in large scale to support more emerging applications. For the purpose of realizing full potential of MEC, it is necessary to allocate computing and caching resources in an intelligent way. The network need to perform decisions adaptively and organize collaboration to further improve the resource utilization, leading to a higher demand for system reliability. Introducing blockchain to the edge network is an effective way to perform resource allocation effectively under the premise of ensuring trustworthiness. To ensure trustworthiness, a credibility evaluation mechanism based on Dempster-Shafer theory is designed in the overlaid blockchain network. Since it is challenging to optimize delay and credibility comprehensively in this dynamic system, we propose a deep reinforcement learning (DRL)-based approach to make offloading decision, and carry out service caching according to both credibility and offloading decisions of multiple users. Finally, the effectiveness of the proposed offloading and caching policies are demonstrated via simulation results.