Precision Sensing-Aided Multi-Target Beamforming Prediction in High-Mobility ISAC Systems Based on OTFS

The integration of orthogonal time frequency space signals into integrated sensing and communication systems has emerged as a highly promising approach for constructing intelligent transportation systems. Among these advancements, beamforming prediction techniques that incorporate multiple-input multiple-output technology have gained widespread attention. However, challenges remain in acquiring initial high-precision state parameters of vehicles and establishing continuous, reliable communication links in high-mobility scenarios. In this paper, we propose a novel MIMO-OTFS beamforming prediction scheme, leveraging a continuous-delay-and-Doppler-shift channel to facilitate information exchange between vehicles and the roadside unit. Furthermore, we develop a three-dimensional parameter estimation algorithm named orthogonal matching pursuit based on maximum likelihood dictionary correction, which offers high accuracy and low complexity. Utilizing this algorithm, we achieve precise multi-target beamforming prediction without approximation. Simulation results demonstrate that our approach significantly outperforms the traditional unscented Kalman filter method based on matched filtering in terms of multi-target beamforming prediction in the ISAC system.