DEI-Department of Electrical and Information Engineering, Politecnico di Bari, 70126 Bari, Italy.
STIIMA-Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing, CNR-Italian National Research Council, 70124 Bari, Italy.
Sensors (Basel). 2019 Apr 6;19(7):1649. doi: 10.3390/s19071649.
Synthetic Aperture RADAR (SAR) is a radar imaging technique in which the relative motion of the sensor is used to synthesize a very long antenna and obtain high spatial resolution. Several algorithms for SAR data-focusing are well established and used by space agencies. Such algorithms are model-based, i.e., the radiometric and geometric information about the specific sensor must be well known, together with the ancillary data information acquired on board the platform. In the development of low-cost and lightweight SAR sensors, to be used in several application fields, the precise mission parameters and the knowledge of all the specific geometric and radiometric information about the sensor might complicate the hardware and software requirements. Despite SAR data processing being a well-established imaging technique, the proposed algorithm aims to exploit the SAR coherent illumination, demonstrating the possibility of extracting the reference functions, both in range and azimuth directions, when a strong point scatterer (either natural or manmade) is present in the scene. The Singular Value Decomposition is used to exploit the inherent redundancy present in the raw data matrix, and phase unwrapping and polynomial fitting are used to reconstruct clean versions of the reference functions. Fairly focused images on both synthetic and real raw data matrices without the knowledge of mission parameters and ancillary data information can be obtained; as a byproduct, azimuth beam pattern and estimates of a few other parameters have been extracted from the raw data itself. In a previous paper, authors introduced a preliminary work dealing with this problem and able to obtain good-quality images, if compared to the standard processing techniques. In this work, the proposed technique is described, and performance parameters are extracted to compare the proposed approach to RD, showing good adherence of the focused images and pulse responses.
合成孔径雷达（SAR）是一种雷达成像技术，其中传感器的相对运动被用来合成一个非常长的天线，并获得高空间分辨率。已经有几种 SAR 数据聚焦算法被空间机构所采用和确立。这些算法是基于模型的，也就是说，必须很好地了解特定传感器的辐射和几何信息，以及在平台上获取的辅助数据信息。在开发用于多个应用领域的低成本、轻量级 SAR 传感器时，精确的任务参数和对传感器所有特定几何和辐射信息的了解可能会使硬件和软件要求变得复杂。尽管 SAR 数据处理是一种成熟的成像技术，但所提出的算法旨在利用 SAR 相干照明，展示了当场景中存在强点散射体（无论是自然的还是人为的）时，从距离和方位两个方向提取参考函数的可能性。奇异值分解被用来利用原始数据矩阵中固有的冗余，相位解缠和多项式拟合被用来重建参考函数的干净版本。即使不知道任务参数和辅助数据信息，也可以从合成和真实原始数据矩阵中获得相当聚焦的图像；作为副产品，从原始数据本身提取了方位波束模式和其他几个参数的估计。在之前的一篇论文中，作者介绍了一项初步工作，该工作能够解决这个问题，并获得高质量的图像，如果与标准处理技术相比的话。在这项工作中，描述了所提出的技术，并提取了性能参数来比较所提出的方法与 RD，显示出聚焦图像和脉冲响应的良好一致性。
