Target Location in WGS-84 Coordinates Using Synthetic Aperture Radar Images

With the advent of GPS/INS integrated navigation for precision guided munitions, terminal accuracies of 15 meters SEP are attainable. The accuracy to which target location is known now can become a significant contributor to the overall error budget. Thus it has become necessary to locate targets in the WGS-84 datum with accuracies significantly less than the 15 meter GPS error. This paper investigates a method for obtaining high accuracy (< 5m) target location using synthetic aperture radar (SAR) imagery of the target. SAB range and range rate equations are solved for target coordinates in the geographic frame, avoiding use of measured angles and coordinate frame transformations that contribute error. Use of one SAR image to solve for target latitude and longitude, and use of two SAR images to solve for target latitude, longitude, and height are discussed. Line of sight velocity errors and their effect on total location error are discussed. Geometric dilution of precision calculations are adapted to the SAR target location problem. Using this information, SAR operators should be able to estimate bounds on the target location performance of their sensor based on navigator performance, SAR motion compensation performance, and radar resolution. This target location method was tested using data from the Sandia National Laboratories Twin Otter radar testbed. A top hat was placed at a known location and imaged, resulting in four data sets comprising over two hundred images. The image sets were processed, yielding target location accuracies ranging from less than 5 meters CEP to 26 meters CEP. Factors which cause this variance in performance are discussed.