基于谐波小波的舰船辐射噪声线谱提取方法

Extraction of Line Spectrum of Ship Radiated Noise Based on Harmonic Wavelet

针对传统方法在低信噪比情况下难以准确提取舰船辐射噪声线谱成分的问题,提出一种基于谐波小波变换的高分辨线谱提取方法。通过对舰船辐射噪声信号进行谐波小波变换,将其正交、无泄漏地分解到相互独立的频段上,提取线谱所在频段的谐波小波系数且将其它频段置零,进而在时域重构出线谱信号,实现信号与其它成分的分离。实验分析结果表明:该方法对背景噪声有较好的抑制作用,提取微弱线谱信号的能力和精度优于FFT分析方法,比传统FIR滤波器方法性能提高了约5dB,有利于辐射噪声特征线谱的提取。

Traditional methods can hardly extract the line-spectrum components of ship-radiated noise accurately in low signal to noise ratio. This paper presented a high resolution method of line-spectrum extraction based on harmonic wavelet transform. By using harmonic wavelet transformation, the ship-radiated noise was decomposed into separate orthogonal frequency bands with no leakage, and the harmonic wavelet coefficients of the frequency bands which contain line spectra was extracted while other bands were set to zero. Then, the line spectra signals was reconstructed in time domain, so that the signals were separated from other components. Experimental results showed that the proposed method was effective in the background noise suppression, and its performance of the weak line-spectrom extraction was superior to Fourier analysis and was about 5 dB higher than FIR filter methods.