On the improvement of channel estimation in transmit-diversity OFDM systems

• Main Authors: Chung-Yu Hsieh, 謝宗佑
• Other Authors: Kuo-Guan Wu
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/59466622479901882265

碩士 === 國立中興大學 === 電機工程學系所 === 102 === Abstract In orthogonal frequency-division multiplexing systems with transmit-diversity,decision-directed channel estimation(DDCE)can be used by cancelling inter-antenna interference to separate each channel estimation. A general conventional methods involve using the least-square algorithm,however,the symbol map can be determine error in DDCE estimation,resulting in decision error,then directly used the outcome of DDCE to perform the least-square estimation can exhibit severe decision error propagation. In this thesis,we proposed an adaptively regularized LS algorithm to improve the performance of DDCE in transmit-diversity OFDM systems.In adaptively regularized LS algorithm,we consider temporal channel correlation,the latest DDCE estimate is adopted as the a priori channel information.and the regularization parameter adaptively determined by the mean square error(MSE)of the original channel and that of the latest DDCE estimate. Index terms- decidion-directed channel estimation(DDCE),adaptively regularized LS(least-square),OFDM.