Research On Correlation Of Massive MIMO Spatial Fading Based On Scattering Cluster Model

The rapid development of the mobile Internet and the Internet of Things has driven the development,adjustment and transformation of wireless communication systems.The fourth-generation mobile communication system has been unable to meet the explosive growth of mobile users and the future needs of the Internet of Things.For low-latency,high-rate and high-band 5G technologies,Massive MIMO technology is the key to break through this bottleneck.This paper mainly studies the spatial fading correlation(SFC),channel capacity and non-stationary characteristics of massive MIMO in different configurations and different scenarios.In this paper,Linear Array(LA),Circular Array(CA),Uniform Plannar Array(UPA)and the classic angular power spectrum distribution PAS(Power Azimuth Spectrum)are used to study it's spatial fading correlation and system performance based on the geometric random scattering cluster model with pitch angle factor introduced.An improved random scattering cluster 3D channel model is builded to analyze the non-stationary characteristics of massive MIMO.The work of this article is mainly divided into the following three aspects:First,based on the Saleh-Valenzuela(SV)scattering cluster model,a two-dimensional planar large-scale MIMO signal reception model is established to mathematically characterize the degree of correlation between the scattered clusters of DOA signal elements in massive MIMO and the spatial fading correlation function of unit distance is obtained.Assuming that the angular power spectrum of the arrival signal under different environments is uniformly distributed,Gaussian distribution,and Laplacian distribution,the closed expression of the spatial fading correlation function of Circular Array(CA)and Linear Array(LA)are drived.At the same time,for the small angle expansion of the arrival signal,an approximation algorithm under three classic PAS is proposed.This paper simulates the system capacity of massive MIMO under different configurations,and compares the performance of the approximate algorithm and accurate analysis.The results are in line with reality and have certain practical significance.Next,in view of the non-horizontal propagation characteristics of electromagnetic waves in urban environments,this paper introduces a pitch angle factor based on the SV scattering cluster model and expands a three-dimensional scattering cluster channel model.For computational efficiency,the Von-Mises distribution instead of the Wrapped Gaussian distribution in the 3GPP model is used and it proves that the two have a similar distribution when the expansion factor in the cluster is small.The analytical expressions of the spatial fading correlation coefficients of the elements of the Uniform Plannar Array(UPA)are derived,and the MIMO multi-antenna system model is established.influences.The simulation results show that the uniform planar array has better performance in the millimeter wave scenario of the cell.Finally,In view of the near-field effect and non-stationary characteristics of massive MIMO channels,this paper proposes a non-stationary three-dimensional spatial channel model of massive MIMO based on random scattering clusters.This paper is proposed to determine the effective scattering cluster set based on the effective probability of the scattering cluster,so as to model the random evolution of the scattering cluster along the antenna array axis to reasonably describe the appearance and disappearance of the scattering cluster.The parabolic wave is used instead of the spherical wave to model the near-field effects.The analytical expressions of the system's spatial fading correlation function and other statistical characteristics are derived to analyzed the channel capacity of the model under the parabolic wavefront.Comparing the simulation data with previous models and data,it is found that the random evolution of effective scattering clusters and parabolic wavefront channel modeling can effectively describe the non-stationary characteristics of massive MIMO.