基于时频分析的海杂波背景下目标检测

发布时间：2018-03-21 18:42

本文选题：海杂波　切入点：IPIX　出处：《哈尔滨工业大学》2014年硕士论文　论文类型：学位论文

【摘要】：海杂波，，来自海洋表面的雷达后向散射回波,海杂波的存在严重干扰了雷达对海面目标的检测和识别能力，因此对海杂波的研究对于雷达系统设计、雷达信号处理和海洋目标检测识别都具有非常重要的意义。目前雷达的海面目标检测技术广泛地应用于海洋安全、海事救生、监控非法走私等领域，具有很高的军事价值和民用价值。论文正是基于上述背景，展开对高分辨率雷达低掠射角下的实测海杂波数据的研究分析。 1)首先研究了海杂波幅度统计特性，针对IPIX相参雷达实测海杂波数据展开幅度统计模型的拟合，并对拟合效果进行拟合优度检验；其次研究了海杂波时间序列、Speckle分量和Texture分量的时间相关性及去相关时间。 2) Hilbert-Huang变换是美籍华裔科学家黄锷在1998年新提出的一种基于信号自身的时频分析方法，主要包括EMD经验模态分解和Hilbert谱分析两个过程。论文中利用相关系数法来去除EMD分解过程中产生的虚假模态函数分量，并在此基础上提出了基于EMD分解的多分量信号WVD变换交叉干扰项抑制方法；然后将Hilbert-Huang变换应用到IPIX相参雷达实测海杂波数据分析处理中，从低频IMF能量比、IMF模态能量熵、Hilbert谱脊线粗糙度三个不同角度实现海杂波背景下的微弱目标检测。 3)基于时间-Doppler速度法和局部幅值统计的思想对IPIX相参雷达实测海杂波数据研究分析，实验结果表明该两种方法是有效的，可以在没有先验知识的条件下检测到目标距离单元的所在。 4)实际海面复杂多变，实测数据往往包含雷达测量噪声和来自粗糙海面的动态环境噪声。为了抑制这些噪声对海杂波数据的内在特性分析的影响，重点研究了基于小波变换的阈值降噪算法，实验仿真表明利用小波阈值法对实测数据进行降噪处理可以取得较好的降噪效果。 5)针对IPIX相参雷达实测海杂波数据处理研究，利用MATLAB软件设计其控制界面GUI，并将各种算法嵌入其中，实现人机交互。
[Abstract]:Sea clutter, radar backscattering echo from ocean surface, the existence of sea clutter seriously interferes the ability of radar to detect and recognize sea surface targets, so the research on sea clutter is designed for radar system. Radar signal processing and ocean target detection and identification are of great significance. At present, radar sea surface target detection technology is widely used in marine safety, maritime rescue, monitoring of illegal smuggling and other fields. It is based on the above background that the research and analysis of the measured sea clutter data under the low grazing angle of high resolution radar is carried out. 1) firstly, the statistical characteristics of sea clutter amplitude are studied, and the fitting of the amplitude statistical model is carried out according to the measured sea clutter data of IPIX coherent radar, and the fitting effect is tested by goodness of fit. Secondly, the time correlation and de-correlation time between speckle component and Texture component in sea clutter time series are studied. 2) Hilbert-Huang transform is a new time-frequency analysis method proposed by Huang E, a Chinese American scientist in 1998, based on the signal itself. It mainly includes two processes: EMD empirical mode decomposition and Hilbert spectrum analysis. In this paper, the false modal function components produced in EMD decomposition are removed by correlation coefficient method. On the basis of this, a method for suppressing cross-interference terms of multi-component signal WVD transform based on EMD decomposition is proposed, and then the Hilbert-Huang transform is applied to the analysis and processing of the measured sea clutter data of IPIX coherent radar. The weak target detection under sea clutter background is realized from three different angles of low frequency IMF energy ratio, IMF mode energy entropy and ridge roughness of Hilbert spectrum. 3) based on the idea of time-Doppler velocity method and local amplitude statistics, the measured sea clutter data of IPIX coherent radar are studied and analyzed. The experimental results show that the two methods are effective. The location of the target distance unit can be detected without prior knowledge. 4) the real sea surface is complex and changeable, and the measured data often contain radar measurement noise and dynamic ambient noise from rough sea surface. In order to restrain the influence of these noises on the inherent characteristic analysis of sea clutter data, This paper focuses on the threshold denoising algorithm based on wavelet transform. The experimental simulation shows that the wavelet threshold method can be used to reduce the noise of the measured data. 5) aiming at the data processing of sea clutter measured by IPIX coherent radar, the control interface of Gui is designed by using MATLAB software, and various algorithms are embedded into it to realize man-machine interaction.
【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】：2014
【分类号】：TN959.72

【参考文献】