On-demand Data Synchronization for Differentiated Digital Twins in Vehicular Networks

In digital twins (DT) enabled vehicular networks, vehicles need to comprehensively consider their different synchronization requirements of data and the resource status of available networks to map parameters to the DTs deployed in the cloud during the driving process. Therefore, how to develop the optimal data synchronization frequency and the ratio of synchronized data volume for each vehicle to maximize its utility has become a key challenge in the data synchronization process. In this article, we propose an on-demand data synchronization scheme for differentiated DTs to obtain the optimal data synchronization strategy between each vehicle and its DT. Specifically, the data synchronization requirement of each vehicle is considered to design a dynamic data synchronization optimization problem for vehicles. Furthermore, the optimization problem is modeled as a Markov decision process and an on-demand data synchronization algorithm based on twin delayed deep deterministic policy gradient (TD3) is designed to obtain the optimal data synchronization frequency and optimal synchronization data ratio for each vehicle. Simulation results show that our scheme can achieve the highest reward compared to conventional schemes.