Efficient Space-Variant Motion Compensation Approach for Ultra-High-Resolution SAR Based on Subswath Processing

Abstract

Motion compensation (MOCO) is essential to obtain high-quality images for airborne synthetic aperture radar (SAR). In the case of ultrahigh resolution, how to compensate the space-variant motion errors accurately and efficiently is still a great challenge. An improved space-variant MOCO scheme is presented in this paper, taking into account both precision and efficiency. By subswath processing, an approximate range envelope compensation without interpolation can be implemented efficiently. Meanwhile, to correct the range-variant component of motion errors effectively, the conventional MOCO processing flow should be modified, guaranteeing that the phase compensation of the one-step MOCO algorithm is not affected by the approximate range envelope compensation. Moreover, based on subswath processing, the range variance of residual azimuth-variant errors can also be degraded, as well as the influence of that on azimuthal time–frequency relationship. Thus, the focusing effect in the case of wide beam is improved significantly, and the computational burden of the precise topography- and aperture-dependent (PTA) algorithm is also reduced. The proposed approach is useful in practice, and the resolution of image corrected by this approach attains 0.1 m. Simulations with point targets and processing of real data have validated the proposed approach.