Research On Consistency Of Millimeter Wave MIMO RF Channels

In mm Wave multiple input multiple output(MIMO)systems,beamforming is a key technology to improve transmission efficiency.However,the components in the radio frequency(RF)channel will inevitably be affected by temperature or hardware aging,and these non-ideal hardware characteristics will affect the performance of beamforming.Thanks to the development of metamaterials,intelligent reflecting surfaces(IRS)technology provides a low-cost and high-efficiency solution to the problem of high-frequency signal occlusion.However,the IRS also faces the problem of suboptimal hardware,which significantly degrades the performance of passive beamforming.Research shows,most hardware impairments or non-ideal properties can be modeled as effects on the phase of the received signal.Therefore,this thesis focuses on the phase calibration technology of mm Wave MIMO systems.In order to study the influence of hardware non-idealities on transmission performance of mm Wave MIMO systems,the hardware impairments of the base station's phase shifter network were modeled,an optimization model for phase estimation was constructed,and a phase calibration algorithm based on an alternate iterative framework was proposed.Aiming at the problem of phase deviation calibration when the channel information is known,a power iterative algorithm based on semidefinite relaxation(SDR)is proposed by row decomposition of the deviation matrix.For the phase calibration problem with unknown channel information,channel estimation and phase estimation can be constructed into three independent sub-problems,and three solutions are proposed: alternate iteration algorithm based on exhaustive method,alternate iteration algorithm based on fast fourier transform(FFT)search,and Alternate iterative algorithm based on estimating signal parameters via rotational invariance techniques(ESPRIT).The simulation results show that the phase calibration algorithm effectively improves the system performance,and the alternate iterative algorithm based on ESPRIT achieves better system gain with low algorithm complexity.Based on the phase calibration of millimeter-wave MIMO systems,this thesis further studies the phase estimation optimization problem of IRS-assisted mm Wave MIMO systems under hardware impairments characteristics,and proposes a joint base station's phase shifter network and IRS phase deviation based on an alternate iterative framework.The algorithm is calibrated and compared with the system CRAMER-RAO lower bound(CRLB)to verify the effectiveness of the algorithm.Aiming at the joint phase calibration problem with known channel information,the phase deviation matrix of the base station phase's shifter network is first decomposed into rows,and the SDR-based power iteration algorithm is used to solve it,and then the SDR-based power iteration algorithm is used to solve the IRS phase deviation again,alternately iterate until convergence; for the joint phase calibration problem with unknown channel information,a 2D-DFT-based signal arrival angle estimation algorithm is proposed,and then the joint estimation of the complex channel gain,the phase deviation matrix of BS phase shift network and IRS is achieved through a similar alternate iterative algorithm.The simulation shows that the phase calibration algorithm has a large performance gain.Finally,the spectral efficiency of the system under non-ideal RF channel conditions is deduced,and the ideal results are compared with the results under non-ideal hardware conditions.The results show that hardware impairments will affect the system performance.After phase compensation,the spectral efficiency of the system is close to the ideal value,which verifies the importance of phase calibration research.The research results of this thesis can provide a theoretical basis and useful reference for the research on radio frequency channel consistency of mm Wave MIMO systems.