Research On Sparse Channel Estimation Algorithms For Massive 3D MIMO Systems

Due to the explosive growth of traffic services in various fields and the ever-increasing demands on the data transmission rate,all these factors have prompted 5G mobile communication systems to conduct further research on system capacity and data transmission rate.Achieving greater bandwidth through the exploitation of under-utilized millimeter wave bands can effectively increase the capacity of the communications network.Massive 3D MIMO technology greatly improves the spectral efficiency and system capacity by deploying a large number of antennas at the base station and developing the vertical spatial freedom of the antenna,by introducing a large amount of spatial freedom.The channel estimation is a very important part of 5G mobile communications.The accuracy of channel estimation is the key to the correct communication.Therefore,the channel estimation algorithm with high design accuracy and good performance has become one of the most basic and important tasks in 5G communication.Since channel estimation is more difficult due to the complexity of the types of noise in the communication environment,corresponding channel estimation algorithms are proposed for massive millimeter wave 3D MIMO channels in Gaussian noise environments and non-Gaussian noise environments respectively.The main research work of this paper is as follows:Firstly,the characteristics of wireless communication channels and the differences between 3D MIMO systems and 2D MIMO systems are researched.The free space loss in wireless communication channels,the decline of wireless channels,various types of noise interference in wireless channels,and their impact on wireless communication are studied.In addition,3D MIMO system and 2D MIMO system are compared and analyzed,and the actual application of 3D MIMO system is given based on the advantages of 3D MIMO system.Secondly,the channel estimation algorithm in the millimeter wave large-scale 3D MIMO system under Gaussian noise environment is studied.The original millimeter-wave large-scale 3D MIMO channel is converted to a beamspace channel by using a lens antenna array that is mathematically equivalent to a spatial discrete Fourier transform matrix.The channel has a sparse structure characteristic.Aiming at the sparse structure of channel,an adaptive sparse channel estimation algorithm is proposed.By comparing the simulation result with the traditional sparse channel estimation algorithm,the proposed algorithm can accurately reconstruct the sparse channel with higher probability than the traditional sparse channel estimation algorithms,and the estimation error is also smaller under the same conditions.Thirdly,we study the channel estimation algorithm in the millimeter wave large-scale 3D MIMO system with non-Gaussian noise satisfying symmetrical ?-stable distribution.Firstly,the sparse channel estimation problem is modeled according to the noise characteristics of Alpha stable distribution and the sparse characteristics of the channel.Then,because the model is not easy to solve directly,we use the alternating iterative method commonly used in the field of optimization to decompose the problem into two subproblems and then iteratively solve them.The proposed algorithm and the traditional sparse channel estimation algorithm are simulated.The simulation results show that the proposed algorithm is obviously superior to the traditional sparse channel estimation algorithm,that is,the proposed algorithm can better reduce the impact of non-Gaussian noise on channel estimation.