AuthorsK. Zhang, J. Cao, S. Maharjan and Y. Zhang
TitleDigital Twin Empowered Content Caching in Social-aware Vehicular Edge Networks
AfilliationCommunication Systems
Project(s)The Center for Resilient Networks and Applications, Simula Metropolitan Center for Digital Engineering
StatusPublished
Publication TypeJournal Article
Year of Publication2021
JournalIEEE Transactions on Computational Social Systems
Volume9
Issue1
Pagination239-251
Date Published04/2021
Publisher IEEE
Abstract

The rapid proliferation of smart vehicles along with the advent of powerful applications bring stringent requirements on massive content delivery. Although vehicular edge caching can facilitate delay-bounded content transmission, constrained storage capacity and limited serving range of an individual cache server as well as highly dynamic topology of vehicular networks may degrade the efficiency of content delivery. To address the problem, in this article, we propose a social-aware vehicular edge caching mechanism that dynamically orchestrates the cache capability of roadside units (RSUs) and smart vehicles according to user preference similarity and service availability. Furthermore, catering to the complexity and variability of vehicular social characteristics, we leverage the digital twin technology to map the edge caching system into virtual space, which facilitates constructing the social relation model. Based on the social model, a new concept of vehicular cache cloud is developed to incorporate the correlation of content storing between multiple cache-enabled vehicles in diverse traffic environments. Then, we propose deep learning empowered optimal caching schemes, jointly considering the social model construction, cache cloud formation, and cache resource allocation. We evaluate the proposed schemes based on real traffic data. Numerical results demonstrate that our edge caching schemes have great advantages in optimizing caching utility.

DOI10.1109/TCSS.2021.3068369
Citation Key28077