Semi-blind Channel Estimation and Equalization by Fast Forward/Backward Approaches for MIMO STBC-OFDM Systems

• Main Authors: Chun-Hsiao Chen, 陳俊孝
• Other Authors: Lena Chang
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/83835464685154850428

碩士 === 國立臺灣海洋大學 === 通訊與導航工程學系 === 99 === This thesis proposed fast semi-blind channel estimation and equalization based on subspace decomposition method for space-time block codes (STBC) and orthogonal frequency-division multiplexing (OFDM) of multiple-input-multiple-output (MIMO) wireless transmission system. First we developed a fast forward/backward(FFB) channel estimation method which uses the structure characteristics of autocorrelation matrix and symmetry property of eigenvectors. The channel estimation is performed through noise subspace decomposition of two half dimensionality sub-matrices obtained from the smoothed forward and backward correlation matrix. The proposed estimation method achieves the same performance as the forward and backward estimation(FBE) method, but only requires one-fourth computation complexity of FBE. Then, the weight coefficients of the equalizer are computed according to the estimated channel parameters. In the study, a signal subspace-based equalizer is proposed based on the orthogonal property of noise subspace and transmission channels, which can reduces the performance degradation of equalizer caused by the channel estimation error. In addition, an equalizer is designed by both the forward and backward receiving data and further simplified by using permutation invariant property for channel matrix. Finally, to further improve the efficiency of equalizer, an averaged equalization method is developed for real-channel environment. Simulation results demonstrate that the proposed signal subspace-based equalizer and the averaged equalization method is superior to the existing least squares-based equalizers.