Resource Allocation for ISAC and HRLLC in UAV-Assisted HSR System With a Hybrid PSO-Genetic Algorithm

With the rapid development of 6G communication and the wide deployment of high-speed rail (HSR), it becomes essential to enhance the utilization of HSR communication resources while ensuring the requirements of communication-sensitive users for high reliability and low latency. Meanwhile, the development of integrated sensing and communication (ISAC), brings more inspiration for smart HSR. In this background, we model an ISAC and hyper-reliable low-latency communication (HRLLC) system for UAV-assisted HSR. We formulate a mixed integer nonlinear programming problem (MINLP) with the objective of maximizing the fair sum rate while satisfying the minimum radar sensing requirement. To solve this problem of nonconvex and high coupling, we propose a hybrid particle swarm optimization-genetic algorithm (PSO-GA) that combines the fast convergence of PSO-only (PSO) and the strong global search ability of GA, with parameter-free penalty functions. Through careful design, PSO-GA dynamically balances the exploration and development capabilities. It achieves the best overall performance with a faster convergence speed than existing algorithms. An average improvement of 29%, 57%, and 42% has been achieved with different numbers of passengers, total transmission power, and number of resource blocks. This article supports the future development of intelligent HSR communication.