Research On Symbol Synchronization And Blind Equalization In Wireless Communication

With the development of information technology,wireless communication is widely used in civil,military and other fields,which promotes the productivity.The majority of research institutions and enterprises are actively engaged in the research of 5g,ultra wideband wireless communication,millimeter wave wireless communication and other new technologies.Meanwhile,some research institutions and scholars continue to improve and perfect the traditional key technologies in wireless communication.The purpose of this paper is to study and improve the symbol synchronization technology and blind equalization technology in wireless communication.Firstly,based on the analysis of the basic principle and mathematical model of the modulation and demodulation of the wireless digital communication system,this paper builds the basic simulation platform of the modulation and demodulation process,including the modules of symbol sequence generation,waveform shaping,IQ modulation,Gaussian noise loading,coherent demodulation,bit error rate calculation,etc.,and realizes the generation and coherent demodulation of 2ASK,BPSK,QPSK,16 QAM modulation signals,through the transceiver The correctness of the simulation program is verified by the analysis of time domain and frequency domain.Fast and accurate symbol synchronization is of great significance to digital wireless communication system.A classical and widely used symbol synchronization method is Gardner algorithm.However,due to the insufficient stopband attenuation ability of interpolation basis function,there is timing accuracy The problem of insufficient timing error.Therefore,this paper improves the interpolation part of Gardner algorithm based on windowing method,studies and develops an interpolation filter with continuously variable tap number,improves the anti aliasing performance of Gardner interpolator,improves the calculation accuracy of Gardner,and develops a timing error estimation method based on the improved raised cosine roll off filter function in the error estimation loop of Gardner This paper improves the problem of large timing error in time offset estimation of fading signals,effectively reduces the timing error after loop convergence,and improves the speed of loop convergence.Based on the above two improvements,the Blackman window aided Gardner(BWGAD)algorithm is developed.Matlab simulation results show that,compared with Gardner algorithm,BWGAD algorithm's time offset estimation value is closer to the theoretical value and has better variance performance,smaller timing error and less iteration times to complete convergence,that is,timing accuracy and algorithm stability are better than Gardner algorithm.Blind equalization is a kind of adaptive equalization method,which is based on the received signal sequence itself without training sequence under the condition of serious channel distortion.Among them,CMA algorithm is a typical blind equalization algorithm,CMA algorithm is very robust,but there are some problems such as convergence speed and convergence accuracy can not be considered.Therefore,based on the cost function and error function of the classical CMA algorithm,this paper develops the blind least mean forth equalization(BLMFE)algorithm.BLMFE algorithm mainly improves CMA in the following two aspects: first,based on the fourth-order error minimization algorithm,an iterative algorithm is designed to solve the high-order tap coefficient of blind equalizer,and the tap coefficient calculation formula in CMA algorithm is replaced.This processing method can make the blind equalizer obtain faster convergence speed and smaller steady-state error.Secondly,based on the gradient descent algorithm,an optimal calculation method is designed to find the optimal step factor,which replaces the constant step size in CMA algorithm.This method improves the calculation accuracy and speed of CMA algorithm.Matlab simulation results show that BLMFE algorithm,compared with the classical CMA algorithm and the logarithm cost function algorithm(LCF),has higher convergence accuracy,smaller steady-state error,lower bit error rate,better constellations and faster convergence speed.