| With the rapid development of the Internet of Things,ultra-high-definition video and autonomous driving technologies,mobile data traffic has grown tremendously.In order to meet the requirements of higher speed,lower latency,lower energy consumption and more reliable multi-scenario communication,the fifth generation(5G)and the 2030-oriented sixth generation(6G)mobile communication systems have become current hot research.As one of the key technologies of mobile communication systems,Massive MIMO technology can effectively improve the spectral efficiency and signal transmission reliability by deploying hundreds of antennas in three-dimensional(3D)space as to fully exploit the spatial freedom of channels.This thesis focuses on Massive MIMO channel modeling and simulation,Massive MIMO spatial angle modeling,Massive MIMO 3D spatial correlation,and favorable propagation conditions of Massive MIMO systems.The specific research contents and innovations are as follows:1.Massive MIMO channel simulation platform implementation and calibration.The wireless channel simulation platform plays a vital role in some scenarios,such as the new technology evaluation of communication systems,wireless network planning,and MIMO OTA test.Based on the 5G channel model standard ITU-R M.2412,MATLAB software is adjusted to build a Massive MIMO channel simulation platform IMT-2020 CM-BUPT with link-level and system-level simulation capabilities.Furthermore,this thesis introduces the design concept of platform modularization and the detailed channel simulation procedures.The simulation platform built here supports flexible MIMO array configuration,0.5-100 GHz simulation frequency band,maximum 10%carrier frequency bandwidth,multi-scenario and multi-state propagation conditions.In addition,in order to ensure the correctness of the outputs of the channel simulation platform,the platform is calibrated for large-scale parameters and small-scale parameters.The large-scale parameter is coupling loss,and the small-scale parameters including delay,power,angle,and channel matrix eigenvalues.The calibration results show that for the same simulation parameters,the outputs of IMT-2020 CM-BUPT are consistent with those of multiple reference platforms,which proves the correctness and availability of the simulation platform.2.Research on multipath angle characteristics of Massive MIMO channel.The angle distribution of a multipath signal characterizes the scatterering of the signal when it propagates in space,which has an important influence on the study of channel spatial correlation.Based on the spatial angle outputs of IMT-2020 CM-BUPT in multiple scenarios,this thesis studies the angle distribution of multipath signals.The results show that the azimuth angle of the multipath signal obeys the wrapped Gaussian distribution and the elevation angle obeys the truncated Laplacian distribution.Besides,due to the scatterers in the vertical dimension are relatively sparse,the value of the elevation angle spread is relatively small,and the angle is more concentrated.In addition,for large-scale antenna arrays with high angle resolution,this thesis has realized a more detailed modeling of the rays in the cluster,that is,a randomized modeling method is used instead of the traditional fixed parameter modeling method.The rays'delay,power,and angle offsets are all randomly generated,which promotes different statistical characteristics between different clusters.And it makes outputs of the simulation platform closer to the real propagation environment.3.Research on 3D spatial correlation of Massive MIMO channel.The spatial correlation(SC)of channels plays an important guiding role in the practical application of MIMO technology,such as spatial multiplexing and beamforming.However,most research on 3D spatial correlation make ideal assumptions about angle distribution,that is,the angle obeys uniform distribution or the angle spread is small.These assumptions are inconsistent with real propagation scenarios.In order to more closely approximate the real propagation environment,this thesis derives a closed-form expression of 3D spatial correlation under actual angle distributions based on the standardized 3D MIMO channel model.The actual angle distribution means the azimuth angle obeys wrapped Gaussian distribution,and the vertical angle obeys the truncated Laplacian distribution.Based on the derived closed-form expressions,this thesis investigates the effects of the angle distribution in different dimensions and antenna arrays on 3D correlation.The results show that larger angle spread and larger antenna spacing can effectively decline the spatial correlation,and the effect of angle distribution in the horizontal dimension on correlation is more obvious than that in the vertical dimension.4.Theoretical analysis of favorable propagation on Massive MIMO channel with actual angle distributions.Favorable propagation(FP)is a key assumption in Massive MIMO systems,which is defined as mutual orthogonality of different users' channel vectors.So far,most research on FP is based on the assumption of ideal Rayleigh channel and only considers the horizontal dimension.However,the actual propagation channel contains both horizontal and vertical dimensions.And the scatterers are hard to obey the uniform distribution,especially for these around the base station.In view of this,based on the standardized 3D MIMO channel model,this thesis theoretically proves the FP conditions under the angle distributions of wrapped Gaussian and truncated Laplacian(WG-TL),as well as Von Mises and truncated Laplacian(VM-TL).Moreover,the FP conditions under uniform distribution have proven to be a special case of the actual angular distributions whose angle spread tends to infinity.Based on the derived expectation and variance expressions of FP conditions,this thesis studies the effects of angle spread in both horizontal and vertical dimensions and antenna array on FP.The results imply that the larger the angle spread,the easier the favorable propagation is to be satisfied.And the effect of the angle spread in the horizontal dimension on FP is more obvious than that in the vertical dimension.Besides,in practical engineering deployment,when physical space is limited,more compact UPA array with half wavelength antennan spacing can be adjusted.Considering that the number of antennas in a Massive MIMO system is always limited,in order to study how close the actual channel is to the favorable propagation channel when the number of antennas is limited,this thesis compares the eigenvalue spread and capacity difference ratio of the channel matrix in different scenarios.The results show that with the same number of antennas,the richer the scattering environment,the closer the actual channel is to the favorable propagation channel,and when the number of antennas is more than 10 times the number of users,the actual channel capacity performance can reach more than 90%of the ideal Rayleigh channel.In summary,for the Massive MIMO technology,a 5G Massive MIMO channel simulation platform is developed and calibrated.Based on the angular parameters in different scenarios generated by the platform,the spatial angle distribution of multipath signals and the modeling of sub-paths within the cluster under a large-scale array are studied.Then,based on the spatial angle distribution of the multipath signal,the 3D spatial correlation under the real angle distribution and the favorable propagation conditions of the Massive MIMO system are derived and analyzed.The effect of angle spread and antenna arrays on it are studied,which lays the foundation for the deployment and application of the MIMO system. |
PDF Full Text Request