A beamforming method based on image tracking and positioning in the LOS scenario

With the antenna number increasing in Massive MIMO systems, overhead used for channel measurement and feedback increases rapidly. When the cellular mobile communication system apply 5GHz or higher frequency, as diffraction ability of beams gets worse, the system becomes more dependent on line of sight (LOS) propagation. However, the widespread use of cameras creates good solutions for image positioning. If the image information of the communication target is determined, combined with the image tracking algorithm, the position of the target in each frame can be calculated. In the case of LOS propagation, the angle information of the LOS path can be directly determined in order to find the best beam direction. In this paper, a beamforming method combined with binocular image tracking and positioning is proposed. Compared with the traditional method, when using this method to find the best codeword, the search steps can be omitted partly and the spectrum efficiency can be improved. By enhancing the beamforming method, it is possible to reduce the beam index error due to the estimated position inaccuracy, thereby further improve the communication quality.