Research On Channel Modeling Of Massive MIMO System

The fifth generation(5G)mobile communication system has gradually begun to use5 G.The performance goals of 5G are high speed,low latency,low power consumption,interconnection of everything and low cost.The first stage of the 5G specification proposed in Release-15 is to apply the pre-commercial deployment.The second phase of the Release-16 plan has been completed in April 2020.The fifth-generation wireless communication system has three widely used key enabling technologies,namely Massive Multiple Iutput Multiple Output(massive MIMO),Heterogeneous Ultra-dense Network and millimeter wave communication(mm Wave).Among them,in addition to increasing network capacity,massive MIMO technology also has the functions of beamforming and cell splitting for specific sectors or specific users,thereby enhancing signal strength and reducing multi-user interference.These advantages over traditional MIMO technology stem from the greater spatial freedom provided by the large antenna elements equipped on the base station(BS).The number of antennas is usually tens or even hundreds.In summary,massive MIMO communication is very important for 5G systems,and it is accurately modeled to facilitate the design,performance evaluation and optimization of potential large-scale antenna array technology algorithms.Is essential.However,large-scale antenna array technology has new propagation characteristics and puts forward new requirements for channel modeling.Therefore,this article innovatively proposes a two-dimensional(2D)random channel model that can accurately describe the propagation characteristics of the massive MIMO channel based on the two new characteristics of the wireless channel of the large-scale antenna array,and finally compares and verifies it with the measured data.This paper introduces in detail the trends and key 5G wireless communications enabling technologies,as well as the evolution of the wireless channel model,and the background and significance of the massive MIMO channel modeling are all described.Next,the evolution process of massive MIMO channel modeling research is briefly described,and then several commonly used channel modeling techniques are described in detail to elicit two typical characteristics of massive MIMO model,namely,spherical wavefront and array non-stationarity.Finally,the existing actual scenario data is compared with the simulation data of our model to verify the correctness of our model.In view of the two new typical characteristics unique to massive MIMO,the traditional MIMO channel models don't fully characterize these two new propagation characteristics.In view of this limitation,this article addresses two shortcomings of the traditional massive MIMO channel model framework:(1)For spherical wavefront modeling,most of the current channel models require the spatial geometric position of the transceiver to meet specific conditions to facilitate related geometric calculations.However,these modeling methods lack universal applicability,because the spatial geometric conditions of traditional models for receiving and sending positions are generally not accurately met in real channel scenarios;(2)The key channel parameters have not yet established a corresponding relationship with the channel measured data results.The existing birth and death process simulation algorithm has high computational complexity.Two aspects of improvement work have been carried out.In response to these two problems,this article has made two improvements.On the one hand,the method of establishing a spatial rectangular coordinate system makes the transceiver no longer need to meet specific geometric constraints;on the other hand,a general Markov process is proposed to model the non-stationarity of the array caused by the appearance and disappearance of clusters,and greatly reduces the computational complexity,and the simulation time is significantly reduced compared with other algorithms.Finally,the simulation data proves the validity of the proposed channel model.In addition,according to the statistical analysis of the external field measurement,various characterizing parameters of the channel model are determined.