【作者】 崔利民；

【导师】 何宜军；

【作者基本信息】 中国科学院研究生院（海洋研究所） ， 物理海洋学， 2010， 博士

【摘要】 X波段雷达能对海表面成像,利用其高时空变化的海杂波图像,并配合相应的反演算法可以获得高分辨率的海浪、表层流等动力学要素。本文结合中国科学院重大仪器装备研制项目“X波段雷达物理参数测量仪”,研究了X波段雷达提取海浪参数和表面流信息的若干问题。本文简要地回顾了X波段雷达海洋遥感的国内外发展历史及研究现状,并介绍了X波段雷达成像原理和调制机制;在此基础上,深入分析了大入射角下不同极化方式雷达海杂波的回波特性。针对不同极化方式雷达具有的不同成像机制,开发了具有独立知识产权的双极化方式X波段雷达波浪监测系统。研究了X波段雷达海杂波提取波浪参数和海洋表层流信息的反演算法,对岸基条件下的X波段雷达成像进行了仿真,对可能影响浪、流参数反演的各种海洋环境因素和雷达系统噪声进行了分析和讨论,并利用仿真图像进行了海浪和海流反演数值实验,分析了影响反演波浪和表面流参数的关键因素。利用双极化X波段雷达进行了为期半年的海上观测,获得了大量的图像序列资料和同步海浪和海流观测资料,利用上述理论和数值模拟形成的X波段雷达浪、流参数反演方法,对所有成像资料进行反演,并与传统观测仪器实测的数据做了对比和分析,获得了令人满意的效果。其中水平极化测量精度有效波高为0.18m,主波波向为21°,平均周期为0.55s,表面流速为0.15m/s,表面流向为21.1°,垂直极化测量精度有效波高为0.18m,主波波向为15°,平均周期为0.49s,表面流速为0.12m/s,表面流向为19.4°。发现垂直极化方式数据在估算海浪参数和表层流速信息略有优势,并证明该系统具有可靠性和实用性,具有重要的应用价值。另外,提出了基于Radon变换反演波浪参数反演的方法。该方法利用海杂波图像的波浪纹理特征,通过对船载雷达的时序图像执行Radon变换,获取了波向、波速、波长和波周期的波浪参数。基于互相关分析技术消除了Radon变换波向探测的180°模糊现象。利用实例反演结果表明Radon变换算法是可行的,与传统方法相比,不仅能满足反演的精度要求而且计算复杂性也要低很多。最后,基于Windows XP平台和Visual C++编译系统开发了“X波段雷达数据采集及物理海洋参数提取”应用软件系统。更多还原

【Abstract】 X-band radar can image sea surface. By analyzing the temporal and spatial structure of the image sequences, the sea surface dynamic parameters related to ocean waves and sea surface currents can be calculated using corresponding inversion algorithm. In this research, which is Instrument Developing Project of the Chinese Academy of Sciences named‘X band radar for ocean physical parameters’, some problems related to retrieving ocean wave and surface current parameters from X band radar image sequences were investigated.In this dissertation, we briefly summarize the history and recent status of the research on ocean remote sensing using X band radar and present the ocean remote sensing principle and imaging mechanism of sea surface by X band radar. Based on this, the characteristics of radar sea echoes for different polarization at high incidence angle are analyzed in detail. Based on the fact that different polarization has different imaging mechanism, a wave and current monitoring system based on dual polarized X band radar, with independent intellectual property rights, is developed.An algorithm to retrieve sea surface current and wave parameters from X band radar images has been developed, and then the numerical simulation of land-based X-band radar sea clutter was done. The relevant ocean environment factors and the radar system noises that influence the observation of the sea surface current and ocean wave are analyzed and discussed. The key factors that influence the inversion accuracy were given by extracting the ocean wave parameters and surface current from the simulated radar image sequences.An experiment lasted six-month is carried out in the North of Yellow Sea using the newly developed X-band dual-polarization radar system. A great number of radar image sequences were obtained simultaneously with observations of ocean wave and current by wave buoy and sea current meter. Following the inversion algorithm described above, ocean wave parameters and sea surface current are retrieved from the collected X band radar images. The results are in a good agreement with these in-situ. It is shown that the standard mean square roots of significant wave height, dominant wave direction, average wave period, sea surface current speed and direction for HH polarization are 0.18m, 21o, 0.55s, 0.15m/s and 21.1o, respectively and For VV polarization, they are 0.18m, 15o, 0.49s, 0.12m/s and 19.4o, respectively. It is shown that VV polarization has the advantage over HH polarization to measure the ocean wave parameters and sea surface current based on our experiment. The reliability and practicability of the remote sensing system of X band radar are also demonstrated.In addition, a Radon transform algorithm for retrieving ocean wave parameters from X band radar images is proposed. First, the textural feature of radar sea clutter images are transferred by Radon transform to retrieve wave parameters, which are ocean wave direction, wave length, velocity and period from radar image series. The inherent 180°ambiguity of wave direction is resolved directly from radar images based on the cross correlation technology. Compared with other analysis method, the new algorithm has high accurate and simple.Finally, a set of software for acquiring the radar data and extracting sea surface dynamics parameters from X band dual-polarized radar system is developed using Visual C++ under Windows XP.更多还原