Maximizing Quality and Minimizing Cost for VCPS in ISAC-Based Vehicular Networks: A Deep Reinforcement Learning Approach

With the continued development of beyond 5G and 6G networks, integrated sensing and communication (ISAC) is emerging as a critical technology to realize various intelligent transportation systems (ITSs), such as vehicular cyber-physical systems (VCPSs). In ISAC-based vehicular networks, logical view constructions based on heterogeneous information sensing and uploading are critical to the realization of VCPS. However, a higher-quality view (i.e., more real-time and accurate) may require more frequent or redundant sensing and uploading. This paper makes the first attempt to maximize the quality and minimize the cost of the VCPS modeling in ISAC-based vehicular networks. We derive an information sensing model based on multi-class M/G/1 priority queue and a data uploading model based on reliability-guaranteed V2I communications. On this basis, we design two metrics, namely, age of view (AoV) and cost of view (CoV). Then, we formulate an optimization problem to maximize the AoV and minimize the CoV. Further, we propose a distributed distributional deep deterministic policy gradient (D4PG) solution implemented in the roadside units (RSUs), which determine the actions on sensing information, frequency, uploading priority, transmission power, and V2I bandwidth jointly. Finally, we build a simulation model and give a comprehensive performance evaluation, and the simulation results conclusively demonstrate the superiority of the proposed solution.