Open AccessFast ISAR autofocus algorithm via sub‐aperture
Author(s) -
Cai Jinjian,
Song Yuanyuan,
Sun Yinghao,
Liu Feifeng
Publication year - 2019
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2019.0409
Subject(s) - autofocus , inverse synthetic aperture radar , computer science , synthetic aperture radar , algorithm , computer vision , motion compensation , aperture (computer memory) , signal (programming language) , artificial intelligence , radar imaging , optics , radar , physics , acoustics , focus (optics) , telecommunications , programming language
Translational motion compensation (TMC) is a significant step for inverse synthetic aperture radar (ISAR), while phase autofocus plays an important role in TMC. By eigenvalue decomposition (EVD), principle eigenvectors of the covariance matrix constructed by the complex range‐aligned signal can be acquired for ISAR‐phase autofocus. Nevertheless, for purpose of enhancing the cross‐range resolution, the observation pulse number should be increased. Large observation pulse number will lead to the heavy computational complexity due to EVD, which may overburden the system. To overcome this issue, a novel sub‐aperture EVD‐based method is proposed here. The full‐aperture signal is divided into multiple partially overlapped sub‐apertures with a novel partition criterion based on the average range profiles, then the conventional EVD‐based methods are utilised to obtain the phase errors in sub‐apertures. After the jump‐phase compensation, combine the phase‐error estimation of all sub‐aperture to complete the full‐aperture phase autofocus. With this method, the computational efficiency can be improved remarkably in typical scenarios, compared with the traditional autofocus algorithms based on EVD. Finally, results for the simulated data and the real measured data are demonstrated to verify the validity of the proposed algorithm.