基于总变分的Curvelet阈值方法在极地冰雷达数据处理中的应用  

The application of total-variation-based Curvelet thresholding for polar ice radar data processing

作　　者：王文鹏[1,2] 赵博[1] 刘小军[1] 

机构地区：[1]中国科学院电磁辐射与探测技术重点实验室 [2]中国科学院大学

出　　处：《电子设计工程》2013年第20期179-182,共4页Electronic Design Engineering

基　　金：国家863计划(2011AA040202);国家自然科学基金(40976114)

摘　　要：基于本实验室极地冰雷达数据处理的需要,本文采用能对雷达数据最优稀疏表示的Curvelet变换,将雷达数据分解为有效信号部分和含噪部分,为了抑制Curvelet阈值引起的振荡和线性失真,采用总变分最小化迭代计算,完成冰雷达数据的降噪。本文中给出了该方法的具体实现步骤,并通过仿真实验验证了本方法的优越性,最后该方法对我国第28次南极科学考察中山站附近地区获取的冰雷达数据进行处理,结果表明该方法在很好抑制噪声的同时,保持了层位边缘信息。A de-noising method is proposed for the processing of polar ice radar data in our laboratory to achieve noise reduction. Curvelet transform, which can represent the radar data in the optimal sparse space, is employed to decompose the radar data into desired signal and noisy data. To reduce the oscillation and linear distortion induced by Curvelet transform, a total variation(TV) minimization technique is adopted. In this paper, the specific implementation steps are presented, and the superiority of this method is verified by simulation. Finally, the proposed method is used to process the ice radar data obtained during the 28st Chinese National Antarctic Research Expedition in the region of Zhongshan station, Antarctica. The result shows that the proposed method can reduce the noise effectively, while preserving the edge of the ice layers.

关 键 词：总变分 阈值 CURVELET 降噪 冰雷达 

分 类 号：TN957.52[电子电信—信号与信息处理]