Abstract：To solve the problem that traditional channel estimation algorithm of orthogonal frequency division multiplexing (OFDM) based on discrete cosine transform (DCT) could not effectively eliminate noise, a new method for channel estimation of OFDM was proposed based on discrete wavelet transform (DWT) and DCT interpolation. The DWT was used to quantify the signal threshold after channel estimation by least square (LS). The average energy of all sampling points within the cyclic prefix (CP) was calculated, which was regarded as threshold in the process of DCT interpolation. The noise in the CP was handled again to further reduce the noise effect. The bit error rate (BER) and the mean squared error (MSE) of the algorithm were simulated. The simulation results show that the improved algorithm is better than traditional LS algorithm, traditional DFT algorithm and DCT algorithm based on improved threshold. The noise inside CP is suppressed effectively by the treatment of new algorithm. Compared with the DCT algorithm based on improved threshold, the SNR of the proposed algorithm is improved by about 1 dB for the same BER and by about 2 dB for the same MSE.
谢斌, 陈琴, 杨丽清. 一种结合DWT与DCT插值的OFDM信道估计新方法[J]. 江苏大学学报（自然科学版）, 2017, 38(3): 313-318.
XIE Bin, Chen-Qin, Yang-Li-Qing. A new method for channel estimation of OFDM
based on DWT and DCT interpolation[J]. Journal of Jiangsu University(Natural Science Eidtion)
, 2017, 38(3): 313-318.
2017, Vol. 38 
