Research On Compensation For Channel Reciprocity In Massive MIMO Systems

With the increasing demand for mobile services,higher spectral efficiency,energy efficiency and system capacity naturally become the goal of the fifth generation of mobile communications(5G).By the virtue of its own antenna number advantage,the large scale MIMO technology provides higher multiplexing gain and diversity gain.Not only does it improve the efficiency of data transmission,but also simplifies the network access mode.Therefore,it has become one of the hotspots for 5G.The numerous advantages of massive MIMO are based on the known channel state information(CSI)in the base station.In the aspect of channel state information acquisition,the TDD system has the inherent advantage of channel reciprocity compared with the FDD system.In the TDD system,the channel reciprocity can simplify the design of the pilot and avoid the resource overhead of the channel state information feedback and the quantization error.However,in massive MIMO systems,the acquisition of channel state information based on channel reciprocity is limited by some factors in practical application,such as the asymmetric characteristics of radio frequency link,the time-varying characteristics of the uplink and downlink channel,and so on.Therefore,it is of great significance to explore the channel reciprocity compensation technology of large scale MIMO channels.This paper comprehensively analyzes the impact factors that affect the application of massive MIMO channel reciprocity,and focuses on the non-reciprocal compensation problem caused by the two key constraints,radio frequency link asymmetry and channel time-varying.The main research work and innovation of this paper can be summarized as follows:(1)In terms of channel reciprocity analysis in massive MIMO systems,this paper firstly describes the principle of channel reciprocity,then summarizes the influencing factors of massive MIMO channel reciprocity,and analyzes the system spectral efficiency for large scale MIMO system with linear precoding in imperfect reciprocity state.(2)As for the research on the loss of massive MIMO channel reciprocity caused by asymmetric radio frequency link,the paper first establishes the channel calibration model based on linear precoding transmission.Based on this model,the system capacity limit is derived by radio frequency link calibration.After the detailed study of the existing massive MIMO radio frequency link calibration algorithms,the advantages and disadvantages of each algorithm are analyzed.Aiming at the shortcomings of these algorithms,a fast packet calibration algorithm is proposed in this paper,which achieves a better compromise between the radio frequency link calibration accuracy and calibration complexity and effectively reduces the system time-frequency resource consumption.(3)In the aspects of the channel reciprocity compensation research based on channel prediction,this paper firstly analyzes the properties of time correlation and spatial correlation in massive MIMO transmission channel.In order to achieve a more accurate channel prediction based on these properties,we investigate the channel prediction algorithm based on space-time filtering.According to the shortcomings of these algorithms,a low complexity space-time serial filtering prediction algorithm is proposed.Relying on the polynomial approximation and the descent iterative principle,the proposed method not only guarantees the prediction accuracy,but also reduces the complexity of the predictor.