Abstract: A block-thresholding wavelet denoising based channel estimation algorithm was proposed. The algorithm improved the channel estimation performance by denoising the channel frequency function estimated by Least Square Estimator using block thrcsholding wavclet denoising. Compared with the conventional algorithms, block thresholding wavelet denoising algorithm ultilized the local correlation of the wavelet coefficients, so using the proposed method the mean square error and symbol error rate can be reduced greatly, and the computation complexity is only linearly proportional to the number effective subcarricrs. hhe proposed method also poses good robustness when the CP length of the system is shorter than the delay spread of the channel. Numeric simulations show the performance improvement of the proposed algorithm to the conventional ones.

Key words: OFDM, Channel estimation, Wavelet denoising, Block thresholding