Recent progress in Bistatic SAR with illuminators of opportunity
Recent progress in Bistatic SAR with illuminators of opportunity
复制标题
DOI：
10.1007/s11431-016-0746-8
复制
发表时间：
2016-11
期刊：
影响因子：
--
通讯作者：
T. Zeng;Tian Zhang;Weiming Tian;Cheng Hu
中科院分区：
文献类型：
--
作者：
T. Zeng;Tian Zhang;Weiming Tian;Cheng Hu
具有机会照明器的双基地SAR（BiSAR）是一种灵活的SAR系统，由非合作发射机和固定或移动接收机组成，如图1所示。近年来，为推动此类系统的发展做出了显着的努力。 Antoniou等人对2015年之前以导航卫星为发射机的BiSAR的进展，包括相关理论和实践进行了很好的回顾。并可参考[1]。本文回顾了2015年和2016年的一些进展，包括新颖的信号处理方法、先进的BiSAR系统配置和潜在有价值的应用，如分布式散射体轮廓提取、局部区域表面变化检测和数字高程模型（DEM）生成等。由于全球导航卫星系统（GNSS）由大量具有不同轨道参数的卫星组成，因此已经构思了利用该功能的想法。曾等人[2]提出了多角度BiSAR成像技术。他们的工作中，选择不同观测角度的北斗二号/北斗二号导航卫星作为照明器，获得相应的BiSAR图像并融合为一幅图像。实验数据中观察到不同角度下目标散射特性造成的图像差异，提出了一种贝叶斯方法对图像进行组合，该方法可以从特定角度的图像中提取一些显着特征，也可以获取一个分布散射体的不同部分，产生具有连续轮廓和特征细节的图像。受不同几何形状导致不同距离和方位分辨率方向这一事实的启发，Santi 等人[3]提出了一种基于多静态采集和特征提取的空间分辨率提高技术。获得不同角度下的BiSAR图像并将其不相干地融合成一幅图像。由于两点目标的旁瓣在交叉位置会产生“鬼”目标，组合图像无法提供照明区域的准确信息。为了解决这个问题，他们提出了一种类似 CLEAN 的临时技术。该技术的特点是迭代处理步骤，其中每个迭代步骤中最亮目标的位置和强度都根据特定几何形状下的点扩散函数（PSF）进行估计，然后可以将该点及其旁瓣一起删除。由于“幽灵”目标仅在存在多个相关目标简档时才会出现，因此前述的迭代估计和去除处理可以确保不存在“幽灵”目标。最后，根据记录的点目标的位置，可以获得点目标的图像。由于发射机和接收机的空间分离，基于全球导航卫星系统（GNSS）的BiSAR面临时间和频率同步问题，这可能导致BiSAR图像散焦。在参考文献[4]中，作者提出了一种综合同步和成像方法来解决该问题。首先估计并消除导航信号的影响，表现为沿合成孔径时间的不规则相位跳跃，以实现预同步处理。此后，成功应用误差抵消原理消除了回波信号的时间和频率误差。最后，可以利用传统的反投影算法来获得聚焦图像。在理论和实践层面上，人们为提高BiSAR图像质量做出了一些重大努力。马等人[5]提出信号带宽组合方案...
Bistatic SAR (BiSAR) with illuminators of opportunity is a kind of flexible SAR system that consists of non-cooperative transmitters and a stationary or moving receiver, as Figure 1 shows. In recent years, notable efforts have been made to promote the development of this kind of system. Progress regarding BiSAR with navigation satellites as transmitters before 2015, including the related theory and practice, has been well reviewed by Antoniou et al. and can be referred from [1]. This paper reviews some of the progress in 2015 and 2016, including novel signal processing approaches, advanced BiSAR system configurations and potentially valuable applications such as distributed scatterers profile extraction, local area surface change detection, and Digital Elevation Model (DEM) generation, etc. As Global Navigation Satellite System (GNSS) consists of a large number of satellites with various orbit parameters, ideas of utilizing this feature have been conceived. Zeng et al.[2] proposed a multi-angle BiSAR imaging technique. In their work, BeiDou-2/Compass-2 navigation satellites with different observing angles were selected as the illuminators and the corresponding BiSAR images were obtained and fused to be one image. Image difference because of target scattering characteristic under various angles was observed in the experimental data and a Bayesian method was proposed to combine the images, which can extract some significant features from images in specific angles and can also obtain different parts of one distributed scatterers, yielding images with continuous profiles as well as featured details. Inspired by the fact that various geometry leads to different range and azimuth resolution directions, Santi et al.[3] proposed a spatial resolution improvement technique based on multi-static acquisition and feature extraction. BiSAR images under different angles were obtained and incoherently fused into one image. As the side-lobes of two points targets could generate “ghost” targets in the intersection positions, the combined image could not provide accurate information of the illuminated region. To resolve this issue, they proposed an ad hoc CLEAN-like technique. This technique features on an iterative processing step, where the position and intensity of the brightest target in every iterative step were estimated based on the point spread function (PSF) under the specific geometry and then the point could be removed along with its side-lobes. As the “ghost” target only appears when more than one related targets profile exists, the aforementioned iterative estimating and removing processing can ensure free of “ghost” target. Finally, the image with point targets could be obtained based on the recorded positions of the point targets. Due to the spatial separation of the transmitter and receiver, Global Navigation Satellite System-based (GNSS-based) BiSAR confronts with time and frequency synchronization problems, which can result in defocusing of BiSAR images. In ref.[4], the authors proposed an integrative synchronization and imaging method to tackle the problems. The effects of the navigation signal, which exhibited as irregular phase jumps along the synthetic aperture time, were firstly estimated and removed to implement pre-synchronization processing. After that, the principle of error neutralization was successfully applied to cancel the time and frequency errors of echo signal. Finally, the traditional back-projection algorithm could be utilized to obtain the focused images.Some significant efforts have been made to improve the BiSAR image quality in both theoretical and practical levels. Ma et al.[5] put forward a signal bandwidth combination scheme …