Abstract:
Aiming at the traditional wavelet threshold function denoising, it can not effectively filter out the specific noise in the signal. Combining the advantages of soft and hard threshold functions, a new threshold function algorithm based on lifting wavelet is proposed to denoise the heart sound signal. Firstly, the heartbeat signal is decomposed by lifting wavelet transform; then the wavelet coefficients are updated by the new threshold function for reconstruction; finally, the Hilbert envelope extraction is performed to improve the denoising effect of the heart sounds after denoising the wavelet soft and hard threshold functions.The experimental results show that the proposed method for improving wavelet denoising has better filtering effect than the soft and hard threshold method, and the extracted curve envelope is more clear and smooth.