舰船辐射噪声低频线谱动态精细特征提取方法研究

Research on dynamic and fine feature extraction method of low frequency line spectrum in ship radiated noise

摘要:

针对舰船辐射噪声低频线谱动态频率精细特征提取存在的频率分辨率低、瞬时频率跟踪性差、频率分析不够精细等问题,提出了一种基于移频频率细化瞬时自相关函数滤波的改进维格纳威利分布(Wigner Ville distribution, WVD)算法Z-FIR-WVD(zoom-finite impulse response filter-WVD)。通过对移频降采样瞬时自相关函数在时域进行FIR(finite impulse response)滤波后的信号进行细化分析,可有效提高对WVD中多线谱交叉项干扰的抑制能力和频率分辨能力,实现对低频线谱动态精细特征的有效提取。仿真实验和实测信号验证结果表明,该算法在保持WVD频率聚集性的前提下,可有效抑制交叉项干扰,并实现对低频线谱的高频率分辨率动态分析,为舰船辐射噪声低频线谱动态频率精细特征提取问题提供了一种新的解决方法。

Abstract:

Addressing the issues of low frequency resolution, poor instantaneous frequency tracking performance and rough frequency analysis in the fine feature extraction of low frequency line spectrum of ship radiated noise signals with dynamic and variable frequency, a modified Wigner-Ville distribution(WVD) algorithm, which is called Z-FIR-WVD (zoom-finite impulse response filter-WVD), was proposed based on optimized filter with frequency-shifted and zoomed instantaneous autocorrelation function. By conducting a detailed analysis of the signal after FIR (finite impulse response) filtering of the instantaneous autocorrelation function of frequency-shifted downsampling in the time domain, the interference suppression ability and frequency resolution of the multilinear spectrum cross terms in the WVD can be effectively improved. This achieves the effective extraction of the dynamic fine features of the low-frequency spectrum. The simulation experiments and actual signal verification results indicate that, while maintaining the frequency aggregation of the WVD, this algorithm can effectively suppress cross-interference terms and achieve high frequency resolution dynamic analysis of low-frequency spectral lines. This offers a new solution for the fine feature extraction of dynamic frequency changes in the low-frequency spectrum of ship radiated noise.