Research On Angle Domain Channel Estimation In Massive MIMO System

With the booming of mobile Internet technology,there is an increasing demand for future high-rate communications.Traditional communication technology cannot meet the huge demand of future communication.Multi-Input Multi-Output(MIMO)technology emerges as the times require.This technology is one of the key technologies of 5G communication.Massive MIMO technology can significantly improve the system's energy efficiency,spectrum utilization and system capacity.The ability to obtain accurate channel state information directly affects the performance of massive MIMO.The number of antennas of the large-scale MIMO system at the base station is huge,the pilot pollution is serious,and channel state information acquisition is difficult.Therefore,it is important to study an effective channel estimation method.Currently,there are two ways to implement massive MIMO system: one is to deploy uniform array at the base station,and the other is to deploy non-uniform array at the base station.Compared with the massive MIMO system with uniform array deployed at the base station,the massive MIMO system with non-uniform array deployed at the base station can greatly reduce the number of antennas,reduce the system complexity,and improve the spectrum efficiency and channel capacity.In this paper,the Angle domain channel estimation method is studied in the context of two massive MIMO systems.Firstly,for the large-scale MIMO system with one-dimensional uniform array deployed on the base station side,this paper proposes a channel estimation scheme based on Butler transform.Firstly,the Butler matrix and MUSIC algorithm are used to obtain the angle information,and then the traditional estimation method is used to obtain the channel gain information,so as to obtain the complete channel state information.Then,for the large-scale MIMO system with one-dimensional non-uniform array deployed on the base station side,this paper proposes a channel estimation scheme based on DFT transform.Firstly,the received data is pre-processed,the processed data is subjected to DFT transform,and the phase rotation is used to estimate the accurate angle information.Then the conventional estimation method is used to obtain the channel gain information,thereby obtaining complete channel state information.Finally,for the large-scale MIMO system with two-dimensional non-uniform array deployed on the base station side,this paper proposes a channel estimation scheme based on 2D-DFT transform.Firstly,the received data is preprocessed,the processed data is subjected to 2D-DFT transform,and the phase rotation is used to estimate the accurate information of the elevation angle and the horizontal angle,and then the conventional channel estimation information is used to obtain the corresponding channel gain information,thereby obtaining complete channel state information.In this paper,simulation experiments are carried out in three cases.The simulation results are improved compared with the traditional LS channel estimation method.