Detection and focus of fast movers within SAR imagery data

Synthetic aperture radar (SAR) is a type of radar imagery that relies both on time and phase data to reconstruct scenes. Because of the sensitivity of phase data in particular, SAR imagery is capable of excellent resolution of static targets but is prone to errors induced by moving targets. The focus of this research is on fast movers within a SAR scene, which typically result in significant smears which degrade image quality. The parameters required to focus on these moving targets allow us to identify certain motion parameters, and in some cases the focus we are able to achieve is enough to allow for target identification of large targets moving up to 100 meters per second. Employing a high-volume MATLAB simulation of spotlight-mode SAR of a constant velocity mover, we apply phase gradient autofocus (PGA) in combination with the first of the Muller-Buffington image-sharpness metrics to determine the validity of the processing parameters used. Since this metric uses the square of the image intensity, it is sensitive to small changes in sharpness. Finding the maximum value of this metric indicates that we have selected the correct processing parameters, and with good enough resolution will give us an indication of the speed and heading of the target. We then modulate the simulated radar platform parameters to determine which will give the best focus on fast targets in an attempt to extend these conclusions to even faster targets.