Digital Twin Empowered Heterogeneous Network Selection in Vehicular Networks With Knowledge Transfer

Digital twin (DT) has emerged as a promising technology to create digital replicas of physical vehicles to facilitate driving in Internet of Vehicles (IoV). However, due to the uneven distribution of vehicles in the networks and the dynamics of heterogeneous wireless links, how to optimize the connection along the vehicle's trip between the vehicle and its DT to achieve the lowest cost (i.e., delay and energy consumption) is a fundamental challenge. In this paper, the authors investigate the joint network selection and power level allocation problem in the data synchronization issue between vehicles and DTs in the heterogeneous access networks. In order to better use the information shared by DT to facilitate the data synchronization process, the authors propose a learning-based heterogeneous network selection scheme in DT empowered IoV. First, the authors consider the data divided into blocks and model the data synchronization problem as a Markov Decision Process (MDP) to minimize the long-term cost in terms of the delay and energy consumption. Based on the information obtained from the digital twin communications, the authors develop an actor-critic learning approach to solve the MDP problem with the block-level state method, and obtain the optimal network selection strategy for the vehicles. In addition, the authors explore the capability of digital twins by applying transfer learning. The vehicles are divided into two types, i.e., learner vehicles and expert vehicles, so that the knowledge can be transferred from the expert vehicles to the learner vehicles to effectively improve the learning efficiency and decrease the data synchronization latency. Extensive experiments demonstrate that the authors' scheme can effectively find the optimal network selection policy and achieve the fast convergence and better performance compared with existing related approaches.

• Record URL:
  http://dx.doi.org/10.1109/TVT.2022.3190941
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/00189545
• Supplemental Notes:
  • Copyright © 2022, IEEE.
• Authors:
  • Zheng, Jinkai
  • Luan, Tom H
  • Hui, Yilong
  • Yin, Zhisheng
  • 0000-0002-3136-788X
  • Cheng, Nan
  • Gao, Longxiang
  • Cai, Lin X
• Publication Date: 2022-11

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: pp 12154-12168
• Serial:
  • IEEE Transactions on Vehicular Technology
  • Volume: 71
  • Issue Number: 11
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 0018-9545
  • Serial URL: http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=25

Subject/Index Terms

• TRT Terms: Data models; Heterogeneity; Knowledge; Motor vehicles; Networks; Wireless communication systems
• Subject Areas: Data and Information Technology; Highways; Vehicles and Equipment;

Filing Info

• Accession Number: 01865811
• Record Type: Publication
• Files: TRIS
• Created Date: Nov 29 2022 9:30AM