Research on Integrated Sensing, Communication Resource Allocation, and Digital Twin Placement Based on Digital Twin in IoV

To address the challenges faced by vehicles using integrated sensing and communication (ISAC) devices in maintaining sensing quality while maximizing the timeliness of sensing information, as well as the allocation of limited edge computing resources when large amounts of sensing data are uploaded to edge servers (ESs), we propose an integrated sensing, communication resource allocation, and digital twin (DT) placement scheme based on DT in IoV. First, a frame structure with an adjustable ISAC slot ratio was designed. Under the constraints of sensing mutual information and uplink communication rate, the timeliness of sensed data is enhanced by minimizing the Age of Information (AoI) during data transmission. Second, a vehicle DT (VDT) placement cost model, incorporating both latency and energy consumption, was constructed. The optimal VDT placement strategy was derived by minimizing placement costs, leading to efficient allocation of edge computing resources. Furthermore, considering the impact of vehicle mobility on real-time interaction between vehicles and VDTs, a migration utility based on vehicle migration latency and migration energy consumption between edge nodes was designed. On this basis, a joint system utility maximization optimization model was established, integrating sensing data AoI utility, VDT placement cost, and migration utility. Due to the continuous and discrete variables in the optimization problem, we develop an algorithm based on multiagent deep reinforcement learning (MADRL) to solve the problem, called SCDTP-MPDQN. Simulation results demonstrate the superiority of the proposed scheme in enhancing sensing information utility and system utility.