Abstract: Wavelet transform is widely used in signal denoising because of its advantages such as low entropy, multi-resolution and de-correlation. Based on the traditional wavelet threshold denoising algorithm, this paper proposes an improved wavelet threshold denoising algorithm based on the existing improved threshold function and threshold selection method. The improved threshold function has better continuity and smaller coefficient deviation; the new threshold selection method combines the Lipschitz characteristic of the signal and has a higher noise signal removal rate. Denoising simulation experiments show that the improved wavelet threshold denoising algorithm improves the signal-to-noise ratio (SNR) and mean square error (MSE) by 14.4% and 58.3% compared with the traditional algorithm, and improves by comparing with existing algorithms 8.4% and 36.5% have better denoising effect, which proves the performance advantages and engineering application value of the proposed denoising algorithm.