基于改进迭代自适应算法的水下近场噪声源定位  

Underwater near-field noise source localization based on improved iterative adaptive algorithm

作　　者：裴秋秋 何琳[1] 帅长庚[1] 胡鹏涛 孙玉臣 PEI Qiuqiu;HE Lin;SHUAI Changgeng;HU Pengtao;SUN Yuchen(Inst.of Noise and Vibration,Naval Univ.of Engineering,Wuhan 430033,China)

机构地区：[1]海军工程大学振动与噪声研究所,武汉430033

出　　处：《海军工程大学学报》2024年第6期66-70,共5页Journal of Naval University of Engineering

基　　金：国家部委基金资助项目(202315F010)。

摘　　要：水下目标在高速移动时,传统的噪声源定位方法受到阵列天线采样快拍数少的影响而精度差。为降低快拍数较少的影响,提高水下噪声源定位的精度,提出了一种基于迭代自适应方法的水下近场噪声源定位方法。首先,构建了近场噪声源接收模型并基于加权最小二乘法建立了该模型的价值函数;然后,对上述模型进行求导得到接收信号最小二乘估计值;最后,将噪声源可能存在的关键位置区域进行了二维网格划分,再运用迭代自适应算法估计每个网格处的信号能量,绘制的能量谱的谱峰位置即为噪声源位置。仿真实验表明:在声源数已知条件下,与高分辨MUSIC方法和MVDR方法相比,所提方法能够对近场噪声源实现高精度定位,尤其在低信噪比和少快拍数下,所提方法也适用于相干噪声源定位。When underwater targets move at high speeds,traditional noise source localization methods are affected by the low number of array antenna sampling snapshots,resulting in poor accuracy.For the high accuracy of underwater sources localization under few snapshots,an underwater near-field noise source localization algorithm was proposed based on iterative adaptive method.Firstly,the value function of the model was established based on the weighted least squares method,according to the established near-field noise source receiving model.And then,the least squares estimation value of the received signal was obtained by derivative of the value function.Finally,the key position area where the noise source may exist was divided into two-dimensional grids,and then the iterative adaptive al-gorithm was adopted to estimate the signal energy at each grid,whereby the energy spectrum was drawn,and the peak position was the noise source position.Simulation experiments show that com-pared with the high-resolution MUSIC algorithm and MVDR algorithm,the proposed algorithm can achieve high-precision of near-field noise source localization,especially in low Rsn and few snapshots,simultaneously suitable for coherent noise source localization.

关 键 词：噪声源定位 快拍数 能量谱 迭代自适应 

分 类 号：TB535[理学—物理]