Research On High Spectral Efficiency Hygbrid Precoding Algotithm For Multiuser Millimeter Wave Maasive MIMO Systems

With the development of economy and information socialization,the transmission rate provided by the fourth generation mobile communication technology has not met the needs of people.In-depth research on 5G key technologies has been launched around the world.Massive MIMO technology can greatly improve the spectrum efficiency of the system and reduce interference between different users.Millimeter wave technology enables high transmission rates by increasing the spectral bandwidth.Therefore,millimeter-wave massive MIMO technology has become one of the key technologies for 5G mobile communication.Since digital precoding methods lead to high cost,high power consumption,and high computational complexity in large-scale MIMO systems,all-digital precoding methods are not practical in large-scale MIMO systems.Therefore,a hybrid precoding algorithm is introduced in a millimeter-wave massive MIMO system.However,existing hybrid precoding algorithms cannot achieve better spectral efficiency performance than traditional digital precoding algorithms.Therefore,the high spectral efficiency hybrid precoding algorithm for multi-user millimeter-wave massive MIMO systems has been deeply studied.The research content of this paper is as follows:In order to further improve the spectral efficiency of multi-user millimeter-wave massive MIMO systems,an improved hybrid block diagonalization precoding algorithm is proposed.Its core idea reconstructs the baseband equivalent channel matrix to increase the dimension of the baseband equivalent channel matrix,which can increase the gain of the baseband equivalent channel matrix,and finally improve the spectral efficiency performance of the whole system.Secondly,an improved matrix decomposition hybrid precoding algorithm based on alternating optimization is proposed.Its core idea is to transform the non-convex problem into a series of convex optimization problems by means of alternating optimization due to the non-convex constraints of the analog matrix.The simulation results show that compared with the traditional algorithm,the improved hybrid block diagonalization precoding algorithm and the matrix decomposition hybrid precoding algorithm based on alternating optimization can achieve better spectral efficiency performance and better bit error rate performance.Therefore,the proposed algorithms are more suitable for millimeter-wave massive MIMO systems.