A Joint Time-Frequency Scaling Algorithm for Squint SAR Imaging

Abstract

Squint radar imaging is an important working mode for synthetic aperture radar (SAR). To facilitate the accumulation of azimuth energy, range walk correction is commonly applied to straighten the envelopes of the scattering points. However, this process violates the principle of translation invariance, causing azimuth spatial-variant effects in range migrations and Doppler coefficients. Consequently, conventional imaging methods struggle to achieve effective and uniform focusing across the entire scene. To address this issue, this paper proposes a joint time-frequency scaling algorithm for squint SAR imaging. First, a keystone transform is employed to correct the spatial-variant range migration, thereby concentrating the energy of scattering points within a single range-sampling unit. Subsequently, variable scaling functions are introduced in both the time and frequency domains to compensate for the spatial variation of Doppler coefficients with azimuth position, effectively improving the azimuth focusing depth and achieving precise focusing of the entire scene. Experimental results validate the effectiveness of the proposed method.