A Noniterative Map-Drift Autofocus Algorithm Using PCA for RDA-Based Stripmap SAR Imaging

• Published in: IEEE sensors journal Vol. 24; no. 16; pp. 25940 - 25948
• Main Authors: Imanifar, Ali, Samadi, Sadegh
• Format: Journal Article
• Language: English
• Published: New York IEEE 15.08.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Apertures; Autofocus; Azimuth; Cost analysis; Drift; Image resolution; map-drift algorithm (MDA); Motion compensation; motion compensation (MoCo); Movement; Navigation; phase gradient autofocus (PGA); Platforms; Principal component analysis; principal component analysis (PCA); Principal components analysis; Radar data; Radar imaging; Radar polarimetry; Remote sensing; Remote sensing systems; Remote sensors; Satellite imagery; Sensors; Spaceborne radar; stripmap; Synthetic aperture radar; synthetic aperture radar (SAR); Weather
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2024.3416197

Synthetic aperture radar (SAR) is a radar remote sensing system that can provide wide-area high-resolution images. One of the prominent features of this sensor is the ability to take images in different weather conditions and during all times of the day and night. This sensor is usually installed on moving platforms such as airplanes and satellites. SAR requires movement without tension and distortion in the required trajectory to achieve a precise and focused image. However, this is never possible in airborne platforms. Therefore, in image formation algorithms, navigation sensor data and motion compensation (MoCo) methods are used to eliminate the effect of this deviation from the ideal path. Achieving the best performance in MoCo requires the use of accurate and expensive navigation sensors. The alternate approach uses ordinary navigation sensors and autofocus methods based on the radar data. In this article, an azimuth direction autofocus method in strip mode SAR is introduced, which is based on Map-Drift for the range-doppler algorithm (RDA). The proposed algorithm uses principal component analysis (PCA) to achieve high accuracy and low processing time. This single-step noniterative method has high accuracy and minimal processing cost. The results obtained from applying the proposed algorithm to simulated and real data show the superiority of this algorithm over similar algorithms.