Low-Latency ISAC Optimization for 6G V2X

Integrated Sensing and Communication (ISAC) is a pivotal technology for 6G vehicle-to-everything (V2X) applications, enabling concurrent target detection and data exchange within a unified framework. Beam misalignment, precoder imperfections, and fluctuating radar cross sections increase sensing latency, challenging the low-latency needs of safety-critical V2X scenarios. In this paper, we develop a robust optimization framework to minimize sensing latency in ISAC systems while preserving high detection probability and communication quality. We propose a novel signal model for antenna-array imperfections and derive analytical detection thresholds and repetition bounds for Unmanned Aerial Vehicle (UAV) and vehicular environments. We optimize sensing repetitions, block length, and detection thresholds via low-complexity Branch-and-Bound for MINLP, achieving 99% lower complexity (0.11 ms) than grid search, which benchmarks optimal results. Simulations show our framework achieves latencies below 15 ms, detection probabilities above 0.95, and robust Signal-to-Interference-plus-Noise Ratio (SINR) despite precoder imperfections, meeting safety-critical V2X demands. The proposed framework thus offers a scalable solution for 6G V2X, effectively balancing sensing and communication performance in dynamic vehicular environments.