Abnormal Noise Detection Method Based on Wavelet Filter and K-L Information

De-noising and extraction of the abnormal noise signature are important to analyze signal in which abnormal noise are often very weak and masked by noise. The wavelet transform has been widely used in signal de-noising due to its extraordinary time-frequency representation capability. In this paper, the wavelet filter-based de-noising methods are introduced to de-noise signals from mechanical defects. In order to select optimal parameters for the wavelet filter, a two-step optimization process is proposed. A periodicity detection method based on singular value decomposition (SVD) and K-L information modeling are used to choose the appropriate scale for the wavelet transform. The experiment result reveals that wavelet filter is more suitable and reliable to detect abnormal noise of mechanical impulse-like defect signals.