基于信号小波系数层间关系的线谱信号去噪研究

Research of the Denoising the line spectrum signal of the Ship-Radiated Noise Based on the Wavelet Inter-scale Dependency

舰船辐射声中的低频线谱是舰船的一个重要特征量,对声引信的信号检测、识别与分类具有重要的作用。随着现代舰船,特别是潜艇声辐射能量迅速降低和海洋环境噪声级逐年增加,利用舰船噪声中线谱信号对目标的发现距离正在减小。本文利用双树复解析小波变换(DT-CWT)对海洋环境噪声和舰船噪声线谱信号进行小波分解,并对小波系数的层间联合分布进行分析,建立了海洋环境噪声和线谱信号的小波系数的层间联合分布的数学模型,并推导出最大后验概率估计子(MAP)的解析表达式,用于去除噪声干扰,检测淹没在海洋噪声背景中的舰船噪声线谱信号。对实测舰船噪声信号和海洋环境噪声的分析表明,所提出的算法能够明显减弱连续谱干扰成分,提高舰船噪声中线谱信号的检测效果。

The low-frequency line spectrum of ship-radiated noise is an important characteristic quantity which plays an important part in the signal detection, recognition and classification of acoustical fuze. With the decreasing of ship, especially submarine radiated acoustic energy and the increasing of ambient noise in the sea, the distance to detect targets based on using the line spectrum signal is getting shorter and shorter year by year. In order to denoise and detect the ship＇s line spectrum signal submerged in ocean noise, a mathematic model of inter-scale joint distribution of wavelet coefficients for the noise and the signal was established, and an analytical expression of the maximum a posteriori （MAP） was deduced, on the basis of decomposing the noise and the signal by the dual tree complex wavelet transform （DT-CWT） and analyzing the inter-scale joint distribution of wavelet coefficients. The denoised results show that the proposed algorithm can evidently weaken the continuous spectrum disturbance component and improve detecting effect of the ship＇s line spectrum signal.