Research On Channel Estimation And Equalization Algorithm For Single-carrier MIMO System

In a wireless communication system,the transmission of signals in the channel is always affected by factors such as multipath effects,resulting in fading and distortion of the received signal.In order to resist channel fading and recover distorted signals,a large number of methods have been proposed.Among them,the single-carrier frequency domain equalization technology obtains the relevant information of the channel and restores the original signal by transforming the received pilot signal into the frequency domain.MIMO technology provides diversity gain for communication systems to combat channel fading through schemes such as space-time coding,combining the two can form a more efficient wireless communication system.The increasing application of machine learning in the field of communication has also brought new inspiration to methods to resist channel fading.The research contents of thesis are as follows:(1)Research on Single Carrier Frequency Domain Equalization(SC?FDE)system.After processing the received pilot signal in the frequency domain,the impulse response information of the channel is obtained,and the original signal is recovered by two equalization algorithms,Zero Force(ZF)and Minimum Mean Square Error(MMSE).The simulation results show that although the ZF algorithm has low complexity,it has a big gap with the MMSE algorithm in terms of equalization performance.(2)In the system of combining single carrier technology with MIMO technology,space-time coding technology is introduced into single carrier MIMO system,and the performance of traditional channel estimation algorithms is studied.At the same time,the channel estimation problem of single carrier MIMO system is solved from the perspective of machine learning.The channel estimation model based on ffdnet is studied and compared with the traditional estimation algorithm.The simulation results show that the performance of the channel estimation method based on ffdnet is significantly improved compared with the traditional estimation algorithm.(3)For the channel equalization problem of single carrier MIMO,three neural network models,FCNN,CNN and perceptionnet,are introduced into the channel equalization method of single carrier MIMO respectively.Three different equalization algorithms are formed,and simulation experiments are carried out on the three equalization algorithms.The simulation results show that the three channel equalization models can effectively improve the reliability of the communication system,and have better BER performance than the traditional MMSE algorithm.Among them,the channel equalization model based on inceptionnet can better process data,and the BER performance is the best.