Research On Low Complexity Signal Detection Algorithm In MIMO Systems

MIMO technology is one of the key technologies in 5G communication.Due to the multi-input and multi-output antenna transmission,the spectrum utilization rate and data transmission rate of the communication system are improved.However,as the number of antennas increases,the interference of signals between different antennas also increases.In order to ensure the reliability of the transmitted signal,a better signal detection algorithm needs to be designed.A good signal detection algorithm is closely related to its detection performance and complexity.Improving the performance of existing signal detection algorithms or reducing the algorithm complexity is of great significance to ensure the transmission of communication systems.This thesis studies the signal detection algorithm in the MIMO system.The main contents of the work are as follows:Firstly,in the traditional signal detection algorithm,the detection effect of the SD algorithm is close to the detection effect of the ML algorithm,but its calculation amount is relatively large.In this regard,some people have studied the SD algorithm combining compression factor control and MMSE idea.Although this algorithm reduces the complexity of the SD algorithm,it increases the bit error rate of the algorithm during 16 QAM modulation.In order to improve the bit error rate of the algorithm,this thesis proposes an improved algorithm based on the algorithm.It uses constant coefficients to adjust the noise in the extended channel matrix,and then uses the SD algorithm to detect the extended channel matrix.The simulation results show that the improved algorithm basically does not affect the complexity of the algorithm during 16 QAM modulation,and reduces the bit error rate of the algorithm.Secondly,in view of the high complexity of the LLL-MMSE algorithm based on shift and amplitude transformation decisions,this thesis proposes an improved algorithm based on the algorithm.It first optimizes the LLL algorithm and uses the optimized LLL to process the channel matrix,and then uses MMSE detection to the processed channel matrix.The simulation results show that the improved algorithm reduces the complexity of the algorithm.At the same time,it has little effect on detection performance.Finally,in order to reduce the error generated by the SQRD-MMSE-PIC algorithm during the detection process,this thesis proposes an improved algorithm combining the ML idea based on the algorithm.According to the ML idea,the improved algorithm can select the detection signal with the smallest euclidean distance.The simulation results show that the improved algorithm improves the detection performance of the algorithm and has little effect on the complexity of the algorithm.