Research On Low Complex Signal Detection Algorithms In MIMO Wireless Communication

MIMO system can improve data transmission diversity and speed in wireless communication.So this technology is widely used in the field of communication and plays an increasingly important role.Massive MIMO(Massive MIMO)system is an extension of MIMO system.On the one hand,it can increase the number of antennas to tens or even hundreds;on the other hand,it can maximize the use of existing resources without increasing the transmission bandwidth to improve channel capacity and spectrum efficiency.Although large-scale MIMO system has such great advantages in wireless communication,it still faces great challenges to promote its use.For example,as the number of antennas increases,the channel matrix dimension is huge and the received signal is more complex.As far as the received signal is concerned,how to correctly restore the real transmitted signal becomes the key to restrict whether MIMO technology can be used in wireless communication.Signal detection,as a very important part of restoring real transmission signals,has also brought new challenges with the emergence of large-scale MIMO systems.Such as Maximum Likelihood(ML),Square Detection(SD),Order Serial Interference Cancellation(OSIC)in large-scale MIMO systems need to solve high computational complexity problems.Zero Force(ZF),Minimum Mean Square Error(MMSE)in large-scale MIMO systems need to solve the problem of poor performance.In this context,in order to ensure the high performance of the signal detection algorithm at the receiver and reduce the complexity of the detection algorithm,this paper makes a deep study of the existing algorithms and proposes an improved scheme.The existing algorithms in this paper focus on Order OSIC and RTS.Firstly,from the point of view of high computational complexity when traditional algorithms are applied to large-scale MIMO systems.This paper deeply studies traditional OSIC algorithms,focusing on the sorting criteria in this algorithm.Knowing that OSIC algorithm needs to calculate a large number of weights in the sorting process,this paper proposes a QR Constellation Constrained Order Serial Interference Cancellation detection algorithm(QR-CC-OSIC)based on QR decomposition.QR-CC-OSIC algorithm combines the idea of QR decomposition and binary classification.After only one simplification and classification,a large number of weight calculations can be avoided,which perfectly solves the problem of high computational complexity of traditional OSIC algorithm.In addition,a new Reactive Tabu Search(RTS)algorithm for large-scale MIMO systems is also studied,focusing on the rules of neighborhood partition and candidate solution selection.Knowing that the computation complexity of RTS algorithm is too high because of too many candidate solutions,this paper proposes a Constellation Constrained Reactive Tabu Search(CC-RTS).CC-RTS algorithm introduces constellation constraints and eliminates unnecessary candidate solutions in traditional RTS algorithm.The optimized RTS algorithm has the same performance as traditional RTS.The simulation results show that the computational complexity of QR-CC-OSIC is much less than that of traditional OSIC algorithm when the performance of QR-CC-OSIC is similar to that of traditional OSIC algorithm.Compared with RTS algorithm,CC-RTS algorithm can also greatly reduce the computational complexity of traditional RTS algorithm when the performance is similar.