Research On Channel Tracking Technology In MmWave Massive MIMO System

Milimeter-wave Massive MIMO is a promising technique in the fifth generation(5G)wireless systems.Millimeter-wave communication provides a larger available spectrum,and it greatly alleviates the spectral resource pressure in the current mobile communication system.Moreover,the millimeter wave(mmWave)with short wavelength is conducive to the integration of large-scale antenna panel.The large path loss for mm Wave radio transmission can be combated through massive MIMO beamforming,which can generate narrow beams with high directivity.Millimeter wave technology and massive MIMO technology can meet the high demand of mobile communication system for system capacity and data rate.Therefore,the millimeter-wave Massive MIMO system has significant research value.This work is supported by the research project "China Unicom Network Technology Research Institute:Research on Key Technology of 5G NR" cooperated with an enterprise.For extracting the channel state information to improve beamforming gain in time-varying environment,this thesis focuses on the research of the channel tracking technology in millimeter-wave Massive MIMO systems.A single-user channel tracking algorithm based on physical channel calibration and a multi-user channel tracking algorithm based on beam space support prediction are proposed,respectively.1)The thesis introduces the basic concept,characteristics and key technologies of mm Wave Massive MIMO.Then we summarize the research status of physical channel tracking and beam space channel tracking in millimeter-wave massive MIMO system.2)For the coupling problem of URAs,the two-stage channel tracking algorithm is proposed,which focuses on obtaining accurate CSI in a single-user massive MIMO system.Firstly,the azimuth and elevation angles are easily estimated based on extending Kalman Filter to obtain the physical channel matrix.Due to the coupling between the azimuth and elevation angles,the EKF algorithm for uniform linear arrays directly extends into a uniform rectangular array with some estimation error.Then,the matrix factorization(MF)algorithm is proposed to calibrate the channel.Simulation results demonstrate that the performance of the proposed algorithm.The proposed channel tracking algorithm can reduce the symbol error rates and increase the tracking time compared with traditional channel tracking algorithms.3)Considering the sparse of millimeter-wave beam space channel,a D-step forward prediction based beam space channel tracking algorithm is proposed,which focuses on improving the accuracy of channel tracking in a multi-user massive MIMO system.Based on the time dependence of LSTM algorithm,the prediction results of the previous D-1 time are used as training parameters to track the beam space channel at the current time.It avoids the estimation error expanding with the increase of continuous tracking time in the process of channel tracking.Aiming at the problem of multi-user interference,an interference elimination scheme is proposed.Based on the beam index,it is determined whether there is interference between users,and then the beam allocation algorithm is defined to redistribute the beam for the interfering user.Simulation results show that the proposed algorithm can effectively reduce the symbol error rates and eliminate interference between multi-users in a slowly-changing environment.