Research On Iterative Detection Algorithm In MIMO System

Since MIMO technology has the characteristics of diversity gain and multiplexing gain,it has attracted wide attention from the scientific and industrial circles.Among them,high performance detection algorithm has always been a key issue in the development of MIMO system.From the optimal maximum likelihood detection algorithm at the beginning to the most commonly used minimum mean square error detection algorithm in the later industry,scholars at home and abroad have been working to find high performance and low complexity MIMO detection algorithm.This paper focuses on the high performance and low complexity iterative detection algorithm based on expectation propagation and belief propagation in MIMO system.This paper discusses the MIMO detection algorithm based on expectation propagation.Firstly,the basic principle of expectation propagation is given,and the corresponding theoretical analysis is given.Then the traditional expectation propagation algorithm is introduced,and the processing algorithm is summarized and analyzed theoretically.The problems in performance and complexity of the traditional expectation propagation MIMO detection algorithm are discussed.To solve these problems,an efficient MIMO detection algorithm based on expectation propagation is proposed,which includes iteration method,distribution replacement update strategy,ordering update of information reliability and spherical auxiliary compression space processing algorithm.Finally,simulation analysis shows that the proposed expectation propagation MIMO detection algorithm has faster convergence speed,lower computational complexity and better detection performance than the traditional expectation propagation detection algorithm under various antenna sizes and modulation orders.Compared with traditional MMSE detection algorithm,efficient expectation propagation detection algorithm also has absolute advantages.Secondly,the iterative detection algorithm of message passing based on factor graph structure is studied and discussed.Firstly,the basic principle of belief propagation is introduced.Secondly,the MIMO detection algorithm of belief propagation for different modulation orders is introduced.The corresponding detection performance is verified by performance analysis and Simulation of the corresponding algorithm.Then,the expectation propagation detection algorithm based on factor graph structure is discussed.The simulation results show that MIMO detection algorithm based on expectation propagation can greatly reduce the computational complexity of MIMO detection without loss of performance.Finally,different iteration detection methods in MIMO-LDPC system are discussed.Firstly,the parallel interference cancellation detection algorithm commonly used at present is given,then the traditional BP-BP joint detection algorithm is discussed,and the corresponding algorithm summary is given.Finally,in order to reduce the iteration computation complexity of MIMO information in the joint detection algorithm,an EPBP hybrid joint detection algorithm is proposed.The simulation results show that this method can greatly reduce the computational complexity of information iteration in MIMO without losing performance compared with BP-BP joint detection.