Study On Algorithm And VLSI Implementation Of MIMO-OFDM Channel Estimation And Equalization

In recent decades,wireless communication affects people's way of life and promotes the rapid social progress deeply.In the same time,people's desire for a better life continues to promote the rapid development of wireless communication in the direction of broadband multimedia.In the new generation of wireless communication,OFDM technology and MIMO technology have become the core technology because of their unique advantages to the wireless channel and led to extensive research in the field of communication.As important parts of MIMO-OFDM system,channel estimation and equalization technology have a direct impact on the transmission quality of the wireless communication system.In this paper,the main study object is MIMO-OFDM channel estimation and equalization based on pilots.The characteristics of time-frequency domain of wireless transmission channel are described firstly.In addition,the principles of MIMO and OFDM technology are introduced.Secondly,based on the analysis of pilot classifications,the pilot-based channel estimation and equalization algorithms are introduced.The channel estimation of data points can be divided into polynomial interpolation estimation,DFT estimation and wiener filter estimation.The equalization algorithms mainly have the zero forcing algorithm and the minimum mean square error algorithm.Then this paper will put forward the improved algorithm of data pilots channel estimation based on wiener filtering and FEC iteration to correct the estimated results.In addition,an improved DFT estimation algorithm based on time-frequency domain windowing is proposed in this paper for suppressing spectral leakage and Gibbs effect in DFT estimation and improving the accuracy of channel estimation.Then MATLAB simulation for comparing the performance of various channel estimations will be done in the TU6 wireless channel model based on COST207 standard.Finally,the VLSI implementation of pilot-based MIMO-OFDM channel estimation and equalization will be completed in Quartus ? platform.In addition,the simulation waveform and hardware resource report are given simultaneously.