Squint InISAR Imaging of Group Targets With Dual-Optimization-Driven Method for True and False Marine Targets Recognition

Abstract

Interferometric inverse synthetic aperture radar (InISAR) imaging possesses the capability to acquire the 3-D geometric structure of targets, thereby holding a natural advantage in countering passive jamming such as the corner reflector and chaff cloud. Nevertheless, the presence of squint scenarios and group targets poses significant challenges to existing InISAR imaging algorithms, and the accuracy of parameter optimization estimation also plays a pivotal role in the quality of 3-D imaging. To tackle these issues, this article proposes a dual-optimization driven InISAR imaging algorithm for squint scenarios and group targets to achieve true and false marine targets recognition. In this technique, a brief and effective squint correction method, the dynamic virtual antenna construction method, is proposed to address the problem of squint scenarios by simplifying the wave path difference with the constant and time-variant terms. In addition, for the parameter estimation, a dual-optimization-driven algorithm (DODA) combining particle swarm optimization and Broyden–Fletcher–Goldfarb–Shanno algorithms is developed, which can enhance the parameter estimation accuracy with equivalent computational efficiency. Furthermore, a group targets imaging approach jointing edge detection and DODA is presented, which integrates image processing and motion compensation to achieve high-quality group targets imaging, thereby enabling the recognition of true and false marine targets. Extensive experiments verify the effectiveness and robustness of the proposed algorithm for InISAR imaging and recognition of group targets at various squint angles and signal-to-noise ratios.