Research and Application of Denoising Method Based on Wavelet Threshold

For effectively detecting the backscattering optical wake flow signal from the noise, the original signal is transferred using four Daubechies wavelet decomposition (db4WD) of one dimension Disperse Wavelet Transformation. This method can distinguish the low and the high frequency section, and the threshold function is used to process the wavelet coefficient. A new threshold function is put forward on the basis of wavelet soft and hard threshold de-noising method. The new function has not only overcome discontinuity disadvantage of the hard threshold function, but also solved constant deviation which exists in the soft threshold function, and uses the entropy as a parameter which can measure the sparsity of the original signal data. The experiment shows that the new threshold function can effectively eliminate noise and offset the disadvantage of soft and hard threshold function. The efficiency of the new function is verified by the experiment.