Application of optimal FrFT order for improving the azimuth resolution of range Doppler imaging algorithm. Issue 4 (19th February 2020)

• Record Type: Journal Article
• Authors: Wang, Zhenli; Wang, Qun
• Is Part Of: IET image processing. Volume 14:Issue 4(2020)
• Page Start: 789
• Page End: 793
• Publication Date: 2020-02-19
• Subjects: Fourier transforms -- radar imaging -- synthetic aperture radar -- spaceborne radar -- Doppler radar -- image resolution -- airborne radar; airborne SAR -- fractional Fourier transform -- high‐resolution RD azimuth imaging algorithm -- synthetic aperture radar imaging -- range Doppler imaging algorithm -- SAR azimuth signals -- optimal FrFT order -- azimuth resolution -- spaceborne SAR measurement data -- azimuth imaging parameters; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2019.1065
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : The traditional range Doppler (RD) imaging algorithm has become the most intuitive and classical method in synthetic aperture radar (SAR) imaging processing because of its easy implementation and high processing efficiency. However, owing to the poor quality of the image produced by RD algorithm, this traditional imaging method is growing increasingly unable to meet practical needs. In an effort to solve the problem of poor azimuth imaging quality, this study proposes a high‐resolution RD azimuth imaging algorithm. The expression of the optimal order of the SAR azimuth signals using fractional Fourier transform (FrFT) is derived in detail herein. A theoretical analysis shows that this optimal order is unique and depends on the azimuth imaging parameters. The experimental results based on airborne SAR simulation data and spaceborne SAR measurement data indicate that the azimuth resolution of the proposed algorithm is improved by 31.6 or 46.8% as against that of the traditional RD algorithm, whereas the peak side lobe ratio (PSLR) and integrated side lobe ratio (ISLR) values obtained via the proposed algorithm are approximately equivalent to those obtained via the traditional RD algorithm.