Optimization Of MIMO Detection Algorithms Based On Constellation Limitation

MIMO technology can multiply the signal transmission rate without using additional transmission energy and bandwidth.The detection technology of MIMO system cancels out channel interference and channel noise by algorithm so as to receive accurate signals at the receiving end.The detection technology of MIMO system has been studied for many years.So far,it can be divided into optimal detection algorithm,linear detection algorithm and non-linear detection algorithm.The optimal detection algorithm has the best detection performance,but its high computational complexity limits its practical application.Linear detection algorithms detect signals through linear filters.The computational complexity of these algorithms is low,but their detection performance is difficult to meet the practical needs.The detection performance of nonlinear detection algorithm is generally due to linear algorithm,but this often increases the computational complexity.So.The tradeoff between detection performance and computational complexity of MIMO detection algorithm has always been the focus of research.The specific contributions of this paper are as follows:First,This paper proposes a signal reliability decision(SRD)structure based on constellation distance.The structure judges whether the output signal is reliable according to the distance between the soft output of the budget method and the constellation point.If it is unreliable,the signal will be further detected.Second,Based on SRD structure,MMSE-SRD algorithm and ML-SRD algorithm are proposed.The MMSE-SRD algorithm only detects part of the unreliable signals by ML algorithm.Compared with the existing ML-PDP algorithm,it reduces the computational complexity of the algorithm on the basis of guaranteeing the invariable detection performance.ML-SRD algorithm greatly reduces the computational complexity of ML algorithm by reducing the number of candidate points of unreliable signals and paying part of the detection performance.Third,Based on the analysis of OSIC algorithm,a hierarchical sorting interference cancellation algorithm,L-OSIC algorithm,is proposed.This algorithm can greatly improve the detection performance of OSIC algorithm by selecting the worst layer for detailed search.Then,based on SRD structure,a selective hierarchical sorting interference cancellation algorithm is proposed,named SL-OSIC algorithm.By reducing the number of candidate points on the worst layer,compared with L-OSIC algorithm,this algorithm significantly reduces the computational complexity with minimal detection performance.Finally,through SRD structure,a parallel interference cancellation algorithm is proposed,which can greatly reduce the computational complexity of OSIC algorithm by eliminating parallel interference.