Research On Decoupling Signal Detection Method Based On MB-SIC In Massive MIMO System

With the development of communication technology,quality-demand for higher transmission rate,lower delay and more enhanced communication accuracy have been requested,which is leading to the concept of 5G communication.Massive MIMO technology has attracted a lot of attention by researchers in 5G wireless communication,because of its capability to enhance spectral effectiveness and channel capacity.Received signal detection method,as one of the key technologies of massive MIMO system,has been the focus of the research for massive MIMO technology.In the massive MIMO system,a large scale of antennas have been installed in base station,which has led to increased computational complexity or poor detection performance.To solve the above problems,the signal model and received signal detection method have been studied based on the distributed massive MIMO system.The mathematic model for distributed antenna system has been introduced on the basis of research for the characteristics of MIMO system;detailed study and profound analysis of the detection method for MIMO system have been given,then the performance has been compared in terms of computational complexity and bit error rate;decoupled signal detection method based on MB-SIC has been proposed and discussed.The main topics for this paper has been summarized below:(1)The signal model for distributed MIMO has been studied:the mathematics model for massive MIMO system has been introduced by studying the characteristics of massive MIMO system and the model of single-user MIMO in details.(2)A low-complexity MMSE-IRC signal detection algorithm based on EVD is studied:first,the conventional detection methods have been elaborated,including linear and non-linear detection methods for the massive MIMO technology has been studied,the linear includes MRC?ZF?MMSE method,while the main study for non-linear goes to the successive interference cancellation detection method,theoretically analyzing the performance respectively and computational complexity;then,by exploiting a dimension-reduction technique based on the EVD method,we have studied a low-complexity MMSE-IRC signal detection algorithm for uplink massive MIMO systems to circumvent the complex high-order matrix inversion;finally,the simulation results show the detection performance of the proposed algorithm and the traditional algorithm.(3)The decoupled signal detection method based on MB-SIC have been proposed and researched:by estimating the performance of sum rat,bit error rate and computational complexity for DSD method,the result is that the proposed method can greatly reduce computational complexity without any loss of the performance.The simulation results show that the DSD based on MB-SIC has a good performance in the sum rate,especially for the situation when there is a large scale of users with increasing number of antenna,the method can reduce the computational complexity,which could put the complicated detector into use for massive MIMO technology.