SAR ground moving target's along-track velocity estimation in the complex image domain via SoWVD

• Published in: Journal of engineering (Stevenage, England) Vol. 2019; no. 20; pp. 7026 - 7029
• Main Authors: Huang, Zuzhen, Ding, Zegang, Meng, Canzhen, Ge, Yinchao
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 01.10.2019; Wiley
• Subjects: along-track velocity estimation; complex image domain; Doppler radar; Doppler rate; frequency modulation; IET International Radar Conference (IRC 2018); linear frequency-modulated signal; object detection; radar clutter; radar detection; radar imaging; radar tracking; SAR ground moving target detection algorithms; SAR image; second-order wigner–ville distribution transform; SoWVD; synthetic aperture radar; synthetic aperture radar image; target tracking; transforms; Wigner distribution; radar tracking; SAR ground moving target detection algorithms; radar detection; along-track velocity estimation; synthetic aperture radar image; transforms; object detection; SoWVD; SAR image; synthetic aperture radar; Doppler rate; radar clutter; target tracking; complex image domain; radar imaging; Doppler radar; second-order Wigner–Ville distribution transform; frequency modulation; Wigner distribution; linear frequency-modulated signal
• ISSN: 2051-3305; 2051-3305
• DOI: 10.1049/joe.2019.0450

A ground moving target is always defocused in the synthetic aperture radar (SAR) image domain due to the along-track velocity. In the image domain, a moving target's complex response along the azimuth direction can be modelled as a linear frequency-modulated signal, therefore, its Doppler rate, as well as the along-track velocity, can be estimated based on the recently proposed second-order Wigner–Ville (SoWVD) distribution transform. The proposed method has a low computational complexity without the search procedure and can be well combined with the SAR ground moving target detection algorithms in image domain, where the signal to clutter-plus-noise ratio is higher than other domains. Finally, the effectiveness of this method is verified with some numerical experiments.