A method of characteristic impulse identification based on Laplace wavelet matching

In order to overcome the influence of white noise and narrow band interference, a method of modal parameter extraction based on Laplace wavelet matching was proposed to identify characteristic impulse response. Firstly, according to property of the characteristic impulse, we should construct Laplace wavelet function dictionary. Secondly, matching pursuit method with waveforms which are selected from a Laplace wavelet dictionary of functions is carried on capture the useful impulse. Besides, focus on the multi-mode mixing phenomena, EMD is introduced to decompose the original mixing signal into several simple IMF components and then IMFs separately are matched with the method proposed in this paper. Simulation and experiments results shows that the Laplace wavelet matching method based on EMD is capable of extracting occurrence time, frequency and damping ratio of the impulse signal and identifying the characteristic impulse automatically.