Improved Wavelet Threshold Denoising Approach for Air Duct Leakage Signal Processing

A threshold function plays an important role in the successful application of wavelet denoising, especially in achieving a better trade-off between denoising performance and signal fidelity. Unfortunately, the existing hard threshold functions have an inherent discontinuities, and the soft threshold functions have a constants bias, which makes it difficult to achieve a balance between signal smoothness and edge preservation in the air duct leakage signal processing. To overcome the challenge, this paper proposes an improved wavelet denoising method based on adaptive thresholding. An adaptive threshold strategy is proposed to overcome the limitations of fixed thresholding methods, and a structural optimization approach is introduced into the soft threshold functions to address the problems of discontinuity and constant bias in traditional thresholding functions. To validate the effectiveness of the proposed method, comparative experiments were conducted on the real-world dataset. The experimental results demonstrate that the proposed method outperforms the five other combinations in terms of extracting valid signal components and suppressing noise. Specially, the time localization error of the proposed method is as low as 0.0039 seconds, indicating its promising application in denoising.