Resource Allocation and Trajectory Planning in Integrated Sensing and Communication Enabled UAV-Assisted Vehicular Network.

This paper investigates the problem of maximizing the average achievable rate in an unmanned aerial vehicle (UAV)-assisted vehicular network, where UAVs and ground base stations (GBSs) jointly serve vehicular users through integrated sensing and communication (ISAC) technology. To balance communication and sensing performance, we maximize the average achievable rate under radar sensing constraints by jointly optimizing UAV trajectory planning, vehicle association, and subchannel allocation. The resulting problem is a challenging mixed-integer nonlinear program (MINLP) due to the strong coupling among decision variables. To address this, we propose an iterative algorithm based on block coordinate descent (BCD), which decomposes the original problem into three subproblems-vehicle association, UAV trajectory planning, and subchannel allocation-by fixing certain variables. These subproblems are solved alternately using successive convex approximation (SCA) and convex optimization techniques. Simulation results verify the effectiveness of the proposed algorithm, demonstrating superior average achievable rate performance compared with conventional methods under radar sensing constraints.