Research On The Noise Reduction Technology For Ls Channel Estimation Algorithm Based On The OFDM System

Nowadays,with the rapid development of wireless communication,OFDM(orthogonal frequency division multiplexing)technology has been widely used in the field of wireless communication because of its favorable performance in frequency spectrum utilization and anti-frequency selective fading.Because of the complex environment of wireless communication,especially due to the issues of unpredictability,time-varying characteristics etc.,the reliability of the OFDM System is severely affected.Therefore,efficient channel estimation techniques are often indispensable.This includes two aspects.On the one hand,channel estimation should be accurate enough;on the other hand,the complexity of channel estimation can not be too high.The channel estimation method of OFDM system is studied in view of these two aspects in this thesis.According to the characteristics of the OFDM system,the channel estimation is generally used in the frequency domain channel estimation method.The commonly used channel estimation algorithm mainly includes LS(Least Square)algorithm,MMSE(Minimum Mean Square Error)algorithm and LMMSE(Least Minimum Mean Square Error)algorithm.Among them,LS algorithm has the lowest complexity,so it is widely used in the OFDM system.But this algorithm also has a serious shortcoming that the impact of noise is relatively large.The noise reduction technology for LS algorithm is studied in the thesis to achieve a higher estimation precision,while the algorithm complexity is also acceptable.Firstly,the transmission characteristics of wireless channel and the channel model used in this thesis are introduced.Secondly,we briefly introduce of the basic principles of OFDM and then the methods of channel estimation are studied in the thesis.It mainly includes the following work: This thesis theoretically analyzes the common algorithms based on pilot-assisted frequency domain channel estimation,including the LS algorithm,the MMSE algorithm and the LMMSE algorithm.Their characteristics and applicability are also analyzed.The improved algorithm based on LS channel estimation are studied,which includes the channel estimation algorithm based on DFT(Discrete Fourier Transform),and the improved LS channel estimation algorithm based on threshold noise reduction.And their advantages and disadvantages are analyzed,and makes their own characteristics moreintuitively by simulation Comparative analysis.Through the summary of the these improved algorithm and the analysis of the power delay distribution of multipath channels,the window function design method based on LS channel estimation is derived.According to the low-pass characteristics of the OFDM system of the virtual subcarrier,this thesis analyzes the time diffusion characteristics of the multipath channel,and proposes a Asymmetric window function design method,so as to ensure the integrity of the channel,the effect of noise is filtered out to the maximum extent,and the accuracy of the channel estimation is improved.By trying different window functions and window width,comparing their performance in LS noise reduction,weighing both the noise reduction and restoration of channel integrity,to find the most appropriate window function and window width.The simulation results show that the improved algorithm proposed in this thesis has a performance improvement of about 2.6d B relative to the LS algorithm.Compared with the traditional symmetric window structure,SNR(Signal to Noise Ratio)optimizes about 0.15 d B performance gain,and the complexity of the algorithm is slightly lower than that of the traditional structure of the symmetrical window,the improved algorithm can achieve good error performance with lower complexity.