面向声呐回波信号的FRI采样技术研究

发布时间：2018-02-02 22:26

本文关键词： 有限新息率声呐回波信号多普勒聚焦目标检测信号恢复　出处：《哈尔滨工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：声呐系统目前正向系统性、综合性方向发展,意味着声呐系统将越来越复杂,有效的采样及数据传输是重要基础。传统奈奎斯特采样产生的数据量对于声呐系统中的水下数据传输及无线数据传输来说过大。因此,研究面向声呐回波信号的高效的采样方案很有必要。近年来提出的有限新息率(Finite Rate of Innovation,FRI)采样理论能够根据远低于奈奎斯特的采样率获得的采样值,利用恢复算法估计出FRI信号的自由参量。本文将声呐回波信号建模为参数化信号后,可以利用有限新息率采样理论来对声呐回波信号进行欠采样。首先对声呐回波信号进行模拟预处理,然后利用低速采样获得的采样值获得声呐回波信号的傅里叶级数系数集,再根据回波信号特点利用不同的恢复算法估计未知参数。主要研究内容包括:1、研究声呐回波信号的建模及FRI采样理论。对声呐回波信号的建模进行了详细的阐述,包括采用亮点模型法对目标回波信号进行建模,采用点散射方法与单元散射方法相结合的方式对混响信号进行建模。研究了FRI采样理论的采样过程及恢复过程,针对不同的采样结构及恢复算法分别进行比较,为利用有限新息率采样理论处理声呐回波信号提供理论基础。2、针对无混响时静止点目标的检测,提出将声呐回波信号建模为由时延和幅度参数决定的参数化信号,能够应用FRI采样理论。分别探究了利用单通道调制低通滤波结构及多通道调制积分结构来获取声呐回波信号的傅里叶级数系数,两种结构有各自的优缺点,可根据具体情况选择适合的采样结构,并分析了对两种采样结构恢复效果的影响因素。3、针对混响背景下的动目标检测,提出将目标回波信号建模为由时延、幅度和多普勒频率参数决定的参数化信号,可以利用FRI采样理论实现欠采样。根据目标回波信号与混响信号多普勒频率不同的特点,应用多普勒聚焦技术对声呐回波信号进行处理,可以排除混响的干扰,先恢复目标的多普勒频率,再恢复时延参数和幅度参数,说明了多普勒聚焦处理的优势并进行了仿真验证。4、声呐回波信号FRI采样的硬件实验验证。完成了硬件部分的模块设计,搭建了由乘法器模块、放大器模块和模拟低通滤波模块等组成的模拟预处理结构。利用LabVIEW上位机编程控制NI PXIe设备完成了信号模拟输出、数据采集、数据存储及上位机重构等功能。本文通过大量实验证明了所设计的FRI欠采样系统能够实现对声呐回波信号的欠采样,压缩比能够达到4.72%。
[Abstract]:Sonar systems are developing in the direction of systematization and comprehensiveness, which means that sonar systems will become more and more complex. Effective sampling and data transmission is an important foundation. Traditional Nyquist sampling produces too much data for underwater data transmission and wireless data transmission in sonar systems. It is necessary to study an efficient sampling scheme for sonar echo signals. A new finite Rate of Innovation has been proposed in recent years. Fri) sampling theory can estimate the free parameters of FRI signal based on the sampling value which is far lower than the Nyquist sampling rate. In this paper, the sonar echo signal is modeled as parameterized signal. The limited new sampling theory can be used to undersample the sonar echo signal. Firstly, the sonar echo signal can be simulated and preprocessed. Then the Fourier series coefficient set of sonar echo signal is obtained by using the sampling value obtained by low-speed sampling, and then the unknown parameters are estimated by different recovery algorithms according to the characteristics of echo signal. The main research contents include: 1. The modeling of sonar echo signal and FRI sampling theory are studied. The modeling of sonar echo signal is described in detail, including the modeling of target echo signal by using bright spot model method. The reverberation signal is modeled by the combination of point scattering method and element scattering method. The sampling process and recovery process of FRI sampling theory are studied. Different sampling structures and recovery algorithms are compared to provide a theoretical basis for processing sonar echo signal using finite new sampling rate theory and to detect static point targets without reverberation. The sonar echo signal is modeled as a parameterized signal determined by delay and amplitude parameters. FRI sampling theory can be applied to explore the use of single-channel modulation low-pass filter structure and multi-channel modulation integration structure to obtain the Fourier series coefficients of sonar echo signal. The two structures have their own advantages and disadvantages. The suitable sampling structure can be selected according to the specific situation, and the factors affecting the recovery effect of the two sampling structures are analyzed. Aiming at the moving target detection under reverberation background, the target echo signal is modeled as time delay. The parametric signals determined by amplitude and Doppler frequency parameters can be undersampled by using FRI sampling theory. According to the characteristics of Doppler frequency of target echo signal and reverberation signal, the Doppler frequency of target echo signal is different from that of reverberation signal. The Doppler focusing technique is used to process sonar echo signal, which can eliminate reverberation interference, recover the Doppler frequency of the target first, and then recover the delay parameters and amplitude parameters. The advantages of Doppler focusing processing are explained and the simulation verification. 4. The hardware experiment of sonar echo signal FRI sampling is carried out. The module design of hardware part is completed and the multiplier module is built. The amplifier module and the analog low-pass filter module are used in the analog preprocessing structure. The LabVIEW upper computer is used to control the NI PXIe equipment to complete the analog output and data acquisition. Data storage and host computer reconfiguration functions. In this paper, a large number of experiments show that the FRI under-sampling system can realize the under-sampling of sonar echo signal, and the compression ratio can reach 4.72.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TB56

【相似文献】