Research On Millimeter Wave MIMO Channel Estimation Algorithm Based On Matrix Filling

Millimeter wave(mm Wave)massive multiple-input multiple-output(MIMO)systems have become a key topic of the research next-generation cellular networks.A switch-based hybrid massive MIMO system has been research hotspots for mm Wave communication because it effectively reduces hardware complexity and improves energy efficiency.Channel state information(CSI)is very important in the massive MIMO wireless communication system,and is directly related to the quality of communication.CSI can be acquired by the channel estimation.Therefore,it is significant for the channel estimation to improve the quality of communication mm Wave massive MIMO systems.In this thesis,we mainly study the channel estimation methods for the mm Wave massive MIMO system based on switches.When acquiring CSI for the hybrid system,there are challenges such as training shortage and high computational complexity.In order to solve the problem that it is difficult to obtain an accurate array response vector when the antenna array is damaged,this thesis proposes a fixed-point continuous approximate channel estimation method based on matrix completion by utilizing the low rank characteristics of the millimeter wave channel.By designing effective precoders and combiners,a small number of received entries of the channel matrix are observed,and then the observations are utilized to estimate the entire channel matrix.By analyzing the performance of the mean square error,the achievable spectrum efficiency,and the computational complexity,the experimental results show that the matrix completion estimator is immune to antenna array impairments.The complexity is also relatively low.In addition,the performance for the line-of-sight(Lo S)channel and the non-line-of-sight(NLo S)channel are compared,which shows that the algorithm we proposed can be utilized in Lo S channel and NLo S channel.It shows that when there are impairments to the antenna array,the matrix completion estimator is not affected by phase and gain errors,and it provides a low-cost and robust solution for millimeter-wave channel estimation.