The parameters estimation and the feature extraction of underwater transient signal

The features of transient signal are extracted to recognize the source. And the ignition signal of the underwater vehicle is a kind of transient signal. After the time-frequency analysis, the ignition signal could be similarly considered as a Linear Frequency Modulated(LFM) signal in a short time. The features of a LFM signal turn to be the initial frequency and the frequency modulation rate. The initial frequency and the frequency modulation rate of the LFM signal is used to be estimated by the traditional slope rate algorithm. But the the traditional method is rough and unautomatical. According to the properties of the LFM, the problems could be solved by the Fractional Fourier Transform(FRFT) algorithm. The discrete scaling transform method is used to extened the continuous FRFT to discrete FRFT. The analysis of the data from experiments of the screw propeller speeding up proved that the FRFT is more effective and accurate than the traditional slope rate method in extracting the feature.