Affected by external noise and various nature disturbances,Wheel Force Transducer (WFT) signal may be completely submerged,\nand the sensitivity and the reliability of measurement can be strongly decreased. In this paper, a new wavelet packet denoising\nmethod based on sparse representation is proposed to remove the noises from WFT signal. In this method, the problem of\nrecovering the noiseless signal is converted into an optimization problem of recovering the sparsity of their wavelet package\ncoefficients, and the wavelet package coefficients of the noiseless signals can be obtained by the augmented Lagrange optimization\nmethod. Then the denoised WFT signal can be reconstructed by wavelet packet reconstruction. The experiments on simulation\nsignal and WFT signal show that the proposed denoising method based on sparse representation is more effective for denoising\nWFT signal than the soft and hard threshold denoising methods.
Loading....