Multi-target Beam Tracking for ISAC-based V2I Assisted Perception

Precise beam alignment is crucial for mitigating the significant propagation loss in millimeter-wave vehicular communication networks. However, conventional beam training methods incur excessive overhead and latency, making them unsuitable for high-mobility scenarios. Existing beam tracking approaches face two key challenges: they require uplink feedback information, which increases system overhead, and they struggle to track targets with complex motion. To address these limitations, this paper proposes an Unscented Kalman Filter (UKF) beam tracking method based on Integrated Sensing and Communication (ISAC) for Vehicle-to-Infrastructure (V2I) scenarios. We design state transitions for multiple motion patterns. Meanwhile, when the radial velocity approaches zero in turning scenarios, we use vehicle feedback information to assist sensing, which not only improves the accuracy of beam tracking but also saves spectrum resources. Simulation results demonstrate that the proposed method can achieve high-precision tracking and accurate beam pointing under various vehicle motion conditions.