Denoising for WFT signal based on sparse representation of wavelet packet coefficient
1.School of Information and Control, Nanjing University of Information Science ＆ Technology, Nanjing, Jiangsu 210044, China; 2.School of Instrument Science ＆ Engineering, Southeast University, Nanjing, Jiangsu 210096, China
Abstract：To solve the problem that the sensitivity and the reliability of measurement were strongly decreased by the influence of various disturbances on wheel force transducer(WFT) signal, according to the compressive sensing theory, a new denoising method was proposed based on sparse representation to remove the noises from WFT signal by recovering the sparseness of wavelet packet detail coefficients. The wavelet packet decomposition was carried out to obtain the detail and approximation coefficients. A framework with constraints was constructed to minimize the number of nonzero coefficients, and the l0 norm problem was boiled down to the l1 norm problem to remove the wavelet packet coefficients of noise by iterative weighted algorithm. The denoised WFT signal was reconstructed based on the resulting wavelet packet coefficients by wavelet packet reconstruction. The experiments were completed to compare the denoising performance of threshold denoising method and the proposed method on WFT signal. The results show that compared with the threshold denoising methods, the proposed method is more effective for improving SNR (signaltonoise ratio) and reserving the detail of the signal.
陈旭, 林国余. 基于小波包系数稀疏表示的WFT信号去噪[J]. 江苏大学学报（自然科学版）, 2014, 35(6): 699-704.
CHEN Xu, LIN Guo-Yu. Denoising for WFT signal based on sparse representation of wavelet packet coefficient[J]. Journal of Jiangsu University(Natural Science Eidtion)
, 2014, 35(6): 699-704.
2014, Vol. 35 
