Research On Beam Tracking And Prediction Technology Of Internet Of Vehicles Oriented To Integrated Sensing And Communication

In recent years,with the continuous evolution of 5G massive input and output antenna array technology and millimeter wave communication technology,the internet of vehicles technology has developed rapidly,and vehicle to infrastructure(V2I)communication has become a new research hotspot.For V2I scenarios,a reliable communication link can only be established when the vehicle is beam-aligned with the RSU.Traditional beamforming and beam alignment require long-term beam training and use pilots to achieve beam tracking,but at the cost of large training overhead and communication overhead.With the gradual development of Integrated Sensing and Communications(ISAC)technology,using ISAC signals to perform predictive beamforming has its advantages.In order to reduce the nonlinear error caused by the traditional tracking algorithm and improve the downlink communication performance at the same time,this paper focuses on the tracking and prediction beam technology of the integration of vehicle networking communication and perception,and proposes two beam tracking prediction algorithms based on the integration of perception and communication.When in a Gaussian nonlinear scene,the text proposes to use the unscented Kalman filter for beam tracking prediction.The unscented Kalman filter effectively avoids the linearization step of the extended Kalman filter,and can more accurately capture the changes in the nonlinear system through the unscented transformation.After verification,the root mean square error of the extended Kalman filter algorithm in Gaussian nonlinear scenarios is twice that of the unscented Kalman filter algorithm,and the downlink communication rate is 20%lower than that of the unscented Kalman filter.In non-Gaussian nonlinear scenarios,this paper proposes a beam tracking prediction algorithm combining unscented Kalman filter and particle filter to improve the beam tracking performance in non-Gaussian scenarios.It is verified by experiments that the algorithm has better performance in both Gaussian and non-Gaussian scenarios,and the root mean square error of the angle of the extended Kalman filter algorithm in the Gaussian scenario is 5 times that of the particle filter algorithm,and the downlink communication rate is higher than that of the particle filter algorithm.Particle filtering is 50%lower.In the Gaussian scene,simulation experiments have verified that when the number of sampling points of the particle filter is much larger than that of the unscented Kalman filter,the root mean square error of the angle of the unscented Kalman filter algorithm is twice that of the particle filter,while the downlink The channel communication rate is 50%lower than that of particle filter.