Nowadays synthetic aperture radar (SAR) imaging systems are used in a wide range of applications. Synthetic aperture radar signal processing is processing radar signals with specific data acquisition geometry, in order to forming high resolution radar images in range and azimuth directions. In order to SAR processing become possible, radar antenna must move over the imaging scene. With antenna motion and SAR signal processing technique, we can generate an analytical antenna with narrow beamwidth in the along-tack direction. In order to make use of the SAR image in different applications, in the first step, it is needed to process the raw signals, acquired by the SAR system, and produce the single look complex (SLC) image. Different SAR signal processing algorithms have been developed to create the SLC image. Range migration algorithm (RMA) is one of the best and popular image formation algorithms from SAR signals, which operates in frequency domain. RMA is mostly used for imaging the Airborne and Spaceborne systems. Furthermore, other capabilities of RMA made it possible to use this algorithm for ground-based SAR (GB-SAR) systems. GB-SAR imaging follows the same imaging geometry principles of airborne and spaceborne systems but, the differences are in shorter analytical antenna in azimuth direction of GB-SAR, look angle of antennas, and targets range to sensor, which are very close, in GB-SAR scenario. In ground based systems, radar sensor is mounted on a straight rail. Radar sensor moves step-wise on the rail and acquires data from the corresponding range profile. At the end of data acquisition process, all the raw signals are stored in a two dimensional array. This two dimensional array in used as input of image formation processor and then the final image will be derived. Range Migration algorithm, consists of four main steps: (1) One dimensional long-track Fourier transform (2) Matched filtering (3) Stolt interpolation (4) Two dimensional inverse Fourier transform in azimuth and along-track directions. In this paper, we are going to develop the RMA for image formation of a Ground-based SAR system, which is used for very close range purposes. Our simulated SAR system operates in S-band and modulates signal form 2.4 GHz to 2.5 GHz. It mounted on a three meters long rail and acquires data every two centimeters. We simulate the different distributions of point targets, such as presenting one point target or presenting nine point targets in the imaging scene, based on the reflected signal model of targets. All the targets’ ranges are less than 40 meters. After simulating acquired two dimensional raw signal of targets, RMA is used to extract the focused targets. In post processing step, Hann window in used to suppress the sidelobes of compressed signals. PSLR and ISLR are used as parameters to analyze the quality of detected targets. The mean PSLR of all examined targets in range direction is -13.1143 dB and in azimuth direction is -13.2153 dB and also the mean ISLR of nine targets in range direction is -5.9726 dB and in azimuth direction is -6.1159 dB. Obtained results are acceptable compared to other imaging modes in higher ranges such as Airborne and Spaceborne imaging systems.
Hosseiny B, Amini J. Evaluation of a Signal Processing Algorithm in a Ground-Based SAR System in Simulated Environment. JGST 2018; 8 (2) :189-198 URL: http://jgst.issgeac.ir/article-1-723-en.html