The widefield polar format algorithm, the Stolt polar algorithm, and the differential Doppler algorithm use variations of a new along-track alignment and formatting system (ATAFS) to generate fine-resolution images from synthetic aperture radar (SAR) data. ATAFS introduces a spatially-variant modification of the SAR phase history storage format to remove the formatting inaccuracies of the conventional polar format algorithm and enable full image quality over large scenes without range curvature distortion or image defocus. These new algorithms are well-suited for processing fine resolution spotlight and ultra-wideband SAR data. Their image quality performance is comparable to that of the range migration algorithm (RMA). Unlike RMA, the new algorithms operate on data stabilized to a fixed reference point to remove the azimuth chirp (the Doppler bandwidth of the reference point) before it compromises processor efficiency.

This paper describes a modified polar format algorithm for SAR image-formation processing. The algorithm focuses raw data assuming a spherical reference surface (ground surface), and, unlike standard polar format algorithms, it assumes spherical rather than planar signal wavefronts. The algorithm, or some variant of it based on the same geometry, is therefore more suitable than the standard polar format algorithm in spaceborne applications for which there is significant curvature of the platform flight track and/or curvature of the Earth surface. The algorithm is described here in the context of spotlight-mode data acquisition.

In synthetic aperture radar (SAR), low scene-contrast may invalidate most of existing autofocus methods, and the limited autofocus performance is also difficult to be verified. Based on a parametric statistical signal model in the coherent processing interval (CPI) of SAR, a novel SAR autofocus method is developed and it is especially applicable to extremely low-contrast scene. Furthermore, the limitation of CPI length and the Cram?r-Rao low bound of autofocus parameter estimation are all analytically obtained. Finally, real measurement data is also exploited to validate the proposed model and the new method.

This paper examines signal processing methods for improving fidelity of backprojection SAR imagery using a preprocessing method that suppresses Doppler aliasing as well as other side lobe artifacts that are introduced by the radar radiation pattern. The algorithm, known as digital spotlighting, imposes a filtering scheme on the azimuth-compressed SAR data, and manipulates the resultant spectral data to achieve a higher PRF to suppress the Doppler aliasing. The merits of the algorithm are studied using the ARL Boom-SAR data.

The use of SAR imaging is an important tool in the laboratory RCS characterization of signature critical platforms. Despite measures to the contrary, air turbulence and mechanical vibration can produce complex perturbations of the target during the imaging process. Model code was written to provide simulations over a wide range of representative target motions and imaging schemes. The slow swept-frequency data collection schemes of many laboratory radars means that the target can undergo significant motion during and between pulses, leading to substantial and time-varying defocusing of range profiles. Conventional motion-correction schemes cannot be used as they rely on the presence of clearly defined range profiles over the imaging process. It was found that replacement of a monotonically increasing waveform with one in which the frequency sampling order was repeatedly randomized could produce a significant recovery of the range profiles. In combination with data averaging, this can provide a significant recovery of the imagery. The pattern of the image degradation was found to have a complex dependence on the radar waveform scheme and target motion characteristics.