Algorithm Research On Channel Estimation In Fast Fading Channel Based On OFDM

With the booming development of a new generation of mobile communication system(eg.4G / 5G),studies of mobile communication technology in high-speed scenarios has has attracted widespread attention.Due to the high speed mobile communication pursuing the goal of large capacity,high rate and low latency,communication processing technology demand is becoming stricter to better meet the validity and reliability of information communication under the high speed scenarios.As one of the key technologies on physical layer of LTE system,orthogonal frequency division multiplexing(OFDM)technology is increasingly concerned by people.This paper studies a basic problem of OFDM system in fast fading channel: research or channel estimation as a key part of OFDM systemsThis paper firstly bases on the analysis on statistical characteristics of wireless channel.And Clarke model,Jakes model,improved Jakes model and TDL model are used respectively to model the wireless channel.Then this paper introduces basis expansion models(BEM)in fast fading channel,OFDM system architecture and the implementation principle as well as the key technology.Then the fast fading channel estimation algorithm based on the BEM is well studied.This article mainly uses the clustering comb pilot in the fast fading channel estimation in OFDM system.Building data model of GCE-BEM channel estimation,the performance of BEM-LMMSE algorithm,BEM-LS algorithm and BEM-BLUE is compared by simulation.Then combining the advantage of BEM-LS algorithm and BEM-BLUE algorithm,the traditional BEM-BLUE algorithm is improved.New scheme adopts the value of BEM-LS estimation as the initial value of the BEM-BLUE iteration,so as to improve the accuracy of the initial iteration value.Especially under the low signal-to-noise ratio,the improved BEM-BLUE algorithm can effectively restrain noise effects,and improve the estimation accuracy of channel parameters.In this article BEM-ISDIC equalization algorithm is applied to the system,improving the traditional BEM-MMSE equalization performance of the algorithm.Finally,the improved algorithm and channel estimation algorithm under different BEM models are carried on the simulation and the performance analysis.In addition,from the elimination of Doppler extension of ICI perspective,some improved GCE-BEM channel estimation algorithms are proposed.First of all,the traditional clustering comb pilot structure is improved.The scheme sets pilot-protection on both sides of the pilot as zero to eliminate data sub-carrier to pilot sub-carrier interference(ICI).As this article refactors channel by adopting the GCE-BEM coefficients,the scheme can enhance the overall performance of channel estimation.Then this article focuses on ICI self-elimination algorithm of channel estimation based on GCE-BEM model.This method takes the traditional time domain hamming filter at the receiving end to realize ICI self-elimination.The receiver squilibriums channel employing the equivalent channel matrix.The improved method turns the frequency channel matrix into a strict banded matrice,so that channel power is concentrated in the main lobe,reducing data sub-carrier interference with pilot sub-carrier.The algorithm also employs Blackman filter with lower sub-carrier coupling coefficient instead of traditional hamming filter,further enhancing the ability to ICI self-elimination.Simulation results show that two kinds of improved schemes can improve the channel estimation performance and reduce the system symbol error rate.