A comparison of spatial sampling techniques enabling first principles modeling of a synthetic aperture RADAR imaging platform

Simulation of synthetic aperture radar (SAR) imagery may be approached in many different ways. One method treats a scene as a radar cross section (RCS) map and simply evaluates the radar equation, convolved with a system impulse response to generate simulated SAR imagery. Another approach treats a scene as a series of primitive geometric shapes, for which a closed form solution for the RCS exists (such as boxes, spheres and cylinders), and sums their contribution at the antenna level by again solving the radar equation. We present a ray-tracing approach to SAR image simulation that treats a scene as a series of arbitrarily shaped facetized objects, each facet potentially having a unique radio frequency optical property and time-varying location and orientation. A particle based approach, as compared to a wave based approach, presents a challenge for maintaining coherency of sampled scene points between pulses that allows the reconstruction of an exploitable image from the modeled complex phase history. We present a series of spatial sampling techniques and their relative success at producing accurate phase history data for simulations of spotlight, stripmap and SAR-GMTI collection scenarios.