Research On Millimeter Wave Hybrid Beamforming Technology For Mobile Cellular Networks

Millimeter wave(mm Wave)communication,which can effectively increase the capacity of the communication system,has been becoming the focus in the industry,because of its abundant spectrum resource.However,due to the high communication frequency,the path loss of millimeter wave signal is very large.The shorter wavelength of millimeter wave makes it possible to integrate more antenna array elements in the same area.Therefore,millimeter wave transmission is conditional to use a larger antenna array to obtain larger array gain,especially to generate narrow beam with high gain and high directivity through beam shaping technology,so as to overcome serious path loss.Considering the complexity,cost,power consumption,it is difficult to achieve full digital beamforming in mm Wave large-scale antenna array system.Instead,a hybrid analog-digital beamforming technology is used to achieve a compromise between hardware cost and performance.This dissertation focuses on the application of millimeter wave large-scale antenna array beamforming technology in mobile millimeter wave communication systems.The main contributions of this dissertation are listed as follows:(1)To address the issue of large overhead in traditional exhaustive beam training algorithm,a hierarchical search method based on hierarchical codebook is adopted in this paper.Three hierarchical beam training algorithms which have the same training overhead are proposed to obtain the best transmit and receive beam pairs based on maximizing the receive signal-to-noise ratio.The performance differences of the three layered beam training algorithms are compared,and the optimal layered beam training algorithm is obtained through simulation analysis.(2)In the existing hierarchical codebook design scheme,the beam gain of the training beam in the beam boundary is low.If the upper beam and the lower beam overlap at the edge of coverage,it will cause the error propagation in the hierarchical beam training process.To address the issue of error propagation,a novel non-aligned hierarchical codebook structure based on the idea of beam rotation is proposed.The beam coverage center area in the lower layer is aligned with the beam coverage edge area in the upper layer,so that the wide beam training errors can be corrected during the next layer narrow beam training process.Numerical results confirm that the proposed scheme can increase the success probability of beam training and is robust to spatial fading.(3)For a mobile communication scenario,small misalignment between the transmit and receive beams may lead to a significant degradation of the received signal quality.This paper proposes a joint optimization of mean square error and extended Kalman filter(EKF)robust beam tracking algorithm.First,the beam angles are predicted with EKF algorithm from the received signal.Secondly,we adopt the minimum mean square error(MMSE)criterion to perform iteratively optimization procedure to eliminate the estimation errors caused by the linear approximation of EKF.Simulation results demonstrate that the proposed algorithm can reduce the angle tracking error rates and increase the tracking time compared with traditional angle tracking algorithms.