水下目标回波与混响的时频形态特征域盲分离
发布时间：2018-12-07 20:39
【摘要】：水下沉底或掩埋目标识别受海底混响干扰严重,并且目标回波与混响在时-频域上均存在混叠,增大了对目标回波与混响的分离难度。根据目标回波与混响的产生机理差异,将图像形态学与时频分析相结合的盲分离算法用于目标回波与混响的分离。推导了目标几何声散射成分与混响在Wigner-Ville时频面的形态特征表达式,利用图像形态学滤波去除Wigner-Ville时频面中的混响及刚性亮点之间交叉项,在时频形态特征域获取解混矩阵,实现了目标回波和混响的分离。仿真与实验数据处理结果表明,结合图像形态学的时频域盲分离算法提高了目标回波信号的信混比。
[Abstract]:Underwater bottom or buried target identification is seriously disturbed by underwater reverberation, and the echo and reverberation of target are overlapped in time-frequency domain, which makes it more difficult to separate target echo from reverberation. According to the difference of the mechanism of target echo and reverberation, the blind separation algorithm combining image morphology and time-frequency analysis is used to separate target echo from reverberation. The morphological expression of the geometric acoustic scattering component and reverberation in the Wigner-Ville time-frequency surface is derived, and the cross terms between the reverberation and the rigid highlights in the Wigner-Ville time-frequency plane are removed by image morphological filtering. The de-mixing matrix is obtained in the time-frequency morphological feature domain, and the target echo and reverberation are separated. Simulation and experimental data processing results show that the time-frequency blind separation algorithm combined with image morphology can improve the signal mixing ratio of the target echo signal.
【作者单位】： 哈尔滨工程大学水声技术重点实验室;哈尔滨工程大学水声工程学院;
【基金】：国家自然科学基金项目(51279033) 黑龙江省自然科学基金项目(F201346)资助
【分类号】：TB56
[Abstract]:Underwater bottom or buried target identification is seriously disturbed by underwater reverberation, and the echo and reverberation of target are overlapped in time-frequency domain, which makes it more difficult to separate target echo from reverberation. According to the difference of the mechanism of target echo and reverberation, the blind separation algorithm combining image morphology and time-frequency analysis is used to separate target echo from reverberation. The morphological expression of the geometric acoustic scattering component and reverberation in the Wigner-Ville time-frequency surface is derived, and the cross terms between the reverberation and the rigid highlights in the Wigner-Ville time-frequency plane are removed by image morphological filtering. The de-mixing matrix is obtained in the time-frequency morphological feature domain, and the target echo and reverberation are separated. Simulation and experimental data processing results show that the time-frequency blind separation algorithm combined with image morphology can improve the signal mixing ratio of the target echo signal.
【作者单位】： 哈尔滨工程大学水声技术重点实验室;哈尔滨工程大学水声工程学院;
【基金】：国家自然科学基金项目(51279033) 黑龙江省自然科学基金项目(F201346)资助
【分类号】：TB56
【相似文献】
相关期刊论文 前10条
1 陈云飞;李桂娟;王振山;张明伟;贾兵;;水中目标回波亮点统计特征研究[J];物理学报;2013年08期
2 李秀坤;张晓云;马涛;;混响和水下目标回波的相位耦合性[J];哈尔滨工程大学学报;2009年07期
3 陈拓;蔡惠智;吴永清;吕曜辉;;基于频域典型相关分析的水下目标回波识别[J];海洋技术;2012年01期
4 姜可宇;蔡志明;胡金华;陆振波;;基于混沌理论的海底混响中目标回波提取(英文)[J];声学技术;2006年06期
5 王强;潘翔;;水下沉底小目标回波的短时FrFT滤波分析[J];浙江大学学报(工学版);2008年06期
6 李金明;赵俊渭;王明洲;，
本文编号：2367828
本文链接：https://www.wllwen.com/guanlilunwen/gongchengguanli/2367828.html
