Research On Ship Target Recognition In Water

There are many kinds of signals in the ocean.The article mainly studies the noise signals of ship targets in water,and identifies the ship target signals in the presence of multiple interference signals in water.Noise signals are generated by underwater targets.The frequency range of noise signals generated by ships in the water is from a few hertz to several thousand hertz.When the signal propagates in seawater,most of the energy in the high-frequency part of the signal is absorbed by the seawater.And the low-frequency part of the signal energy is not easily absorbed and reflected by seawater,and can be transmitted over long distances.Therefore,this paper uses the low-frequency part of the noise signal to identify the ship target in the water.Aiming at the frequency characteristics of the ship’s target signal,a wavelet filter that meets the frequency characteristics of the low-frequency part of the ship’s target signal is constructed based on a half-band filter that satisfies both the orthogonal condition and the complete reconstruction condition.A wavelet filter is used to construct the wavelet of the ship’s target signal,and a wavelet library is established that matches the frequency characteristics of the ship’s target signal.Through the wavelet library of ship target signals,the waveform characteristics of simulated ship target signals are extracted.The wavelet library is used to correlate the sample data to obtain the waveform correlation coefficient,and the correlation coefficient value is used as the signal waveform recognition feature value.According to the time-frequency analysis of the signal,the frequency,duration and energy characteristic values of the signal are extracted,and the signal waveform characteristic values are combined to construct a training sample data set of the signal characteristics.The signal feature sample data set is trained by the least square method to obtain the weight vector of the classifier.The classifier recognizes the simulated ship target signal,analyzes the recognition result of the signal,and recognizes the simulated ship target signal.Finally,according to the experimental data of the ship’s target signal collected by the experiment,the waveform characteristic value,frequency,duration and energy characteristic value of the experimental signal are extracted,and the ship target noise signal is identified by combining with the classifier.The experimental results show that the method used in this paper has achieved a good recognition effect on ship targets in water.