Identification of Cabin Noise Source Based on Wavelet Transform

Abstract

In order to solve the problem of error and abstract in the identification process caused by current wavelet methods using grayscale to identify noise sources, a cabin noise source identification method based on wavelet transform(WT) was proposed. Firstly, multi-wavelet reconstruction signals were constructed from original time-domain signals through Matlab Wavelet Toolbox. Secondly, the eigenvectors were defined by calculating the energy value of reconstructed signals at all levels. And the correlation coefficients between cabin noise signal and other signals were obtained by these eigenvectors. Then the extent of vibration or noise energy radiation influence on the cabin noise was estimated by the size of correlation coefficients. Finally, an example was also given to show the effectiveness of the method.