Collaborative Caching in Vehicular Edge Network Assisted by Cell-Free Massive MIMO
Wang Chaowei, Ziye Wang, Xu Lexi, Xiaofei Yu, Zhang Zhi, Weidong Wang
Abstract
The 6G mobile communications demand lower content delivery latency and higher quality of service for vehicular edge network. With the popularity of content-centric networks, mobile users are paying more and more attention to the delay and reliability of fetching cached content. For reducing communication costs, increasing network capacity and improving the content delivery, we propose a collaborative caching scheme based on deep reinforcement learning for vehicular edge network assisted by cell-free massive multiple-input multiple-output (MIMO) system, in which the macro base station is considered as the central processor unit, and the roadside units are treated as roadside access points (RSAPs). The proposed scheme can effectively cache contents in edge nodes, i.e., RSAPs and vehicles with caching capability. We jointly consider the mobility of vehicles and the content request preferences of users, then we use deep Q-networks algorithm to optimize the caching decisions. Simulation results show that the proposed scheme can significantly reduce the content delivery average latency and increase the content cache hit ratio.