Research On Channel Estimation And Tracking In Millimeter Wave MIMO Systems Based On Bayesian And Kalman Algorithms

In recent years,with the rapid growth of mobile data traffic and the constant consumption of communication system frequency resources,improving the system performance has gradually become one of the important goals of the development of mobile communication.In the process of wireless system signal transmission,the channel in the wireless communication is complex and invisible.And the fading and interference between the base station and the terminal have a great impact on the wireless channel,so it is meaningful to study the estimation and tracking of the wireless channel.Channel estimation and tracking is based on the received signal and the system channel model to describe the characteristics of the wireless channel.Generally,channel parameters are used as the estimation and tracking objects.In this paper,the research of channel estimation is based on massive MIMO system,and the tracking of dynamic channel is based on millimeter wave massive MIMO system.Time-dependent block Bayes algorithm based on compressed sensing theory(TMSBB-CS)and Extended Kalman Filtering channel tracking algorithm(EKFCTA)are proposed respectively.The goal is to improve the estimation accuracy and tracking performance.The specific research contents are as follows:(1)In order to improve the spatial freedom and throughput of the system,a massive MIMO technology is explored to deploy a high-dimensional antenna array in the base station.The extended Saleh Valenzuela channel model is used to describe the millimeter wave massive MIMO system channel,and TMSBB-CS is used to estimate the sparseness of the massive MIMO channel.In order to be more practical,EKFCTA is used to track the dynamic channel,and the error range between the measured angle and the real angle is 0 to 5.(2)The channel information in massive MIMO system is a high-dimensional matrix,so the computation will be greatly increased and the processing method will become complex.Therefore,it is necessary to reduce the dimension.Because the number of scatters in the transmission channel is limited,the channel information is sparse.In this paper,a block Bayesian algorithm is proposed,which can obtain the prior information of the channel and estimate the channel by using the timedependent sparsity of the channel.This time-dependent block Bayesian algorithm(TMSBB-CS)can greatly reduce the pilot overhead.Moreover,and the mean square error(MSE)value of TMSBB-CS algorithm has little difference under different time correlation,so the performance of TMSBB-CS algorithm is stable under the condition of time correlation.(3)When the user position changes with time or relative to the base station,the communication quality will become very poor.In order to solve this problem,firstly,the extended Saleh Valenzuela model is used to represent massive MIMO channel of millimeter wave.Secondly,according to the characteristics of the channel,an extended Kalman filter algorithm is proposed to track the massive MIMO channel of millimeter wave.And the path is predicted according to the physical change rule of users.Finally,the performance simulation is carried out.The experimental results show that the extended Kalman filter tracking algorithm can achieve the same tracking performance compared with other algorithms with only a lower signal to noise ratio(SNR).