부산대학교 도서관

Vehicle-to-infrastructure (V2I) scenarios commonly demand both high-accuracy positioning and high-rate communication. In this study, we consider the base station equipped with an extremely large lens antenna array (ExLAA) that serves the vehicles located in the near field region. To overcome the challenges posed by the “window” type array response, we propose a novel direction-of-arrival (DoA) estimation based on sparse Bayesian inference (SBI). In this method, we treat the distance factor in window function as a nuisance parameter which is estimated by the classical matched filtering (MF) due to the sensing ability. Moreover, we develop an extended Kalman filter (EKF)-based closed-loop tracking strategy to assist communication links. This scheme fully exploits the beam tracking capabilities and its energy-focusing characteristics of ExLAA, resulting in a substantial reduction in communication overhead for high mobility scenarios. Numerical results demonstrate that our approach achieves good parameter estimation performance while establishing stable communication links for mobile vehicles.