Seamless and Flexible Factorised Processor for Long-range UAV-SAR

Abstract:
This work describes a seamless and versatile factorized focusing procedure dedicated to large-scale, high-resolution Synthetic Aperture Radar (SAR) imaging. The algorithm should deal with complex trajectories typical of Unmanned Aerial Vehicles (UAV) and should be efficient enough to quickly and accurately focus large images at high resolution. Two scenarios are tackled. The former is a ground-based-like SAR with a small aperture and a wide area covered by a large antenna aperture. The latter is proper stripmap scenario, with large trajectories and wide areas. The entire processing scheme is based on a modified version of the Fast Factorized Back Projection (FFBP), exploiting an ad-hoc reference system in which the resolution is constant, and the number of samples in the back-projection grid is minimal. The developed processor defines the best reference system and the strategies for the minimum computational cost. The theoretical performances of the entire workflow are derived as a rough number of operations required to focus an image. In addition, the results are supported by a numerical simulation comparing the different computational times of each algorithm in the ground-based-like UAV SAR scenario.