Accurate Wide Angle SAR Imaging Based on LS-CS-Residual.

Area of Science:

• Radar Imaging
• Signal Processing
• Electromagnetics

Background:

• Wide Angle Synthetic Aperture Radar (WASAR) systems acquire data over large angular ranges, leading to aspect-dependent scattering phenomena.
• Existing imaging methods, such as L1 regularization-based compressed sensing (CS), struggle with accurate recovery of these aspect-dependent scattering characteristics.
• Subaperture images in WASAR can be conceptualized as frames in a dynamic sequence, exhibiting significant overlap in their support sets.

Purpose of the Study:

• To introduce and evaluate the Least Squares on Compressed Sensing Residuals (LS-CS-Residuals) method for enhanced WASAR imaging.
• To improve the accuracy of aspect-dependent scattering recovery in WASAR imagery.
• To address the computational challenges associated with traditional CS methods in WASAR imaging.

Main Methods:

• The study adapts the LS-CS-Residuals algorithm, originally designed for time-varying sparse signals, to the WASAR imaging problem.
• Azimuth-range decoupled operators are employed within the LS-CS-Residuals iterative process to mitigate high memory demands.
• The method leverages the overlapping support sets between WASAR subaperture images (akin to SAR video frames).

Main Results:

• The LS-CS-Residuals method demonstrates superior accuracy in estimating aspect-dependent scatterings compared to conventional CS-based approaches.
• Real data processing validates the effectiveness of the proposed method in WASAR imaging.
• The use of azimuth-range decoupled operators successfully reduces the computational memory footprint.

Conclusions:

• LS-CS-Residuals offers a more accurate approach to recovering aspect-dependent scattering in WASAR imagery.
• The adapted method provides a computationally efficient solution for WASAR imaging challenges.
• This technique enhances the fidelity of WASAR imaging by addressing inherent scattering variations.