项目名称: 声纳混响的分类特征获取与包络统计分布估计
项目编号: No.61471352
项目类型: 面上项目
立项/批准年度: 2015
项目学科: 无线电电子学、电信技术
项目作者: 马晓川
作者单位: 中国科学院声学研究所
项目金额: 90万元
中文摘要: 实现混响噪声背景下的可靠的目标检测,是声纳工程领域亟待解决的重要课题。主动声纳检测的混响包络背景噪声具有非高斯统计分布特性。混响包络统计分布与混响散射源的物理性质、声纳波束宽度、声传播路径与距离等都有密切的关系。传播过程中海底、海面、水体的反向散射强度不同且经过多途叠加。传统上接收系统加工混响形成的包络会用一个粗略的概率分布统计模型描述,导致声纳在混响背景下检测目标的能力很弱。更好的混响包络生成方法、更精确的随传播过程中混响体物理模型和声纳照射区域变化的混响包络统计模型,可以减少背景统计方差、改善检测能力。本项目结合水下航行器平面阵设计,深入研究混响数据特征量与混响包络统计分布模型的对应规律,以实现混响数据的统计建模与分类。特别强调了对信道变化稳健的混响方位谱特征的获取方法研究和对混响分类性能改善方法的研究。本项目提供实现混响包络数据快速模型分类和在线模型参数估计的理论途径和方法,将促进主动声纳在混响背景下的检测能力提高。
中文关键词: 混响包络统计建模;混响体物理特性及分类;方位谱特征获取;水下航行器声纳恒虚警检测;阵列信号处理
英文摘要: Target Detection of active sonar in reverberation-limited environment is a challenging task due to the high fluctuations of reverberation power. Recent researches show that the reverberation envelope distributions depend on the physical parameters of scatters,the beamwidth of sonar and the propagation path and range. So the surface,the sea bed and the inhomogeneous structure of the sea will backscatter the reverberations in different envelope distributions and the distributions will also change with array designs and observation ranges. This proposal suggests classification of the reverberation data coming from different kinds of scatters that may suggest a lower-variance distribution modeling the reverberation envelope in each class. The researches on reverberation statistics,reverberation classification features will be conducted. And the features such as high-resolution spatial spectrum feature are emphasized due to their potential robustness to propagation. Experiments in plan are designed for AUV sonar that searching the sea in a mobile way.
英文关键词: modeling of reverberation envelope probability density functions;classification of oceanic scatters;high-resolution spatial spectrum of reverberation;CFD in AUV active sonar;array signal processing