Synthetic aperture radar imaging with fractional Fourier transform and channel equalization

This paper investigates the Range-Doppler Algorithm based on the Fractional Fourier Transform (RDA-FrFT) to obtain High-Resolution (HR) images for targets in Synthetic Aperture Radar (SAR) imaging. A mathematical framework for the RDA-FrFT is developed in this paper with closed-form expressions for the range and azimuth compression. The channel effect is considered in this paper for the first time with three inverse techniques to reduce this effect; inverse filter deconvolution, Linear Minimum Mean Square Error (LMMSE) deconvolution, and regularized deconvolution. The performance of the RDA-FrFT is compared with the classical RDA, which is based on the Fourier Transform (FT). Simulation results reveal that the RDA-FrFT offers better focusing capabilities and greater side-lobe reduction ratios. The reflectivity profile obtained with the RDA-FrFT demonstrates a superior performance to the classical RDA. Results show also that the RDA-FrFT gives low Peak Side-Lobe (PSL) and Integrated Side-Lobe (ISL) levels after range and azimuth compression for the detected targets. Finally, the results reveal that the proposed regularized deconvolution technique enhances the performance of the RDA-FrFT significantly if the channel effect is considered.