舰船水声信号特征提取方法研究

Research on Feature Extraction Method of Ship Acoustic Signal

信号特征提取是目标识别的关键技术,为实现水中目标信号的识别,提出一种利用小波库提取水中舰船目标信号波形特征的方法。该方法针对水中舰船目标的水声信号及其频率特性,在满足理想重建条件下,构造正交的小波滤波器,进而构造符合舰船水声频率特性的小波库。利用小波库将信号分解,结合最小二乘法提取舰船信号的波形特征值,并借助Matlab仿真工具进行仿真。结果表明:用小波库方法提取的信号波形特征可降低信号识别维度,在一定的误差范围内识别水中舰船目标信号。

Signal feature extraction is the key technology of target recognition.To achieve the recognition of target signals in water,a method for extracting the waveform characteristics of waterborne ship target signals using wavelet library is proposed.The method is based on the underwater acoustic signal and its frequency characteristics of the ship target in the water.Under the ideal reconstruction condition,the orthogonal wavelet filter is constructed to achieve a wavelet library that accords with the underwater acoustic frequency characteristics of the ship.The wavelet library is used to decompose the signal,and the waveform eigenvalue of the ship signal is extracted by the least squares method.The simulation is carried out by means of MATLAB simulation tool.The results show that signal waveform features extracted by the wavelet library method can reduce the signal recognition dimension and identify the target signal of the ship in a certain error range.