Bistatic synthetic aperture imaging of moving targets

Abstract

This thesis studies imaging of moving targets in a bistatic SAR (Bi-SAR) configuration. Conventional SAR systems and associated image formation methods are optimized to image stationary scenes. The imaging of moving targets is a challenging task since these targets appear smeared in SAR images. Additionally, clutter due to stationary scatterers may occlude the relatively weak scattered signal from moving targets.

In the second part of the thesis, a theory is introduced to quantify the artifacts due to moving targets in Bi-SAR images. Next, this theory is extended to a two-channel Bi-SAR configuration to quantitatively analyze the moving target artifacts and to determine moving target signatures in displaced phase center antenna images. The results of the analysis are used to develop a novel method for the detection and velocity estimation of moving targets embedded in stationary clutter. The theory and methods are validated using simulated and real data.

The first part of this thesis presents theory, methods and algorithms for Bi-SAR imaging of moving targets. Starting from first principle, it presents and analyzes a bistatic received signal model from a moving scene. Next, it introduces a novel filtered backprojection type imaging method to reconstruct two-dimensional reflectivity images and an entropy optimization-based method for velocity estimation. The point spread function of the imaging operator is studied and a two-stage filter is designed to recover the correct scene reflectivity and to optimize the velocity estimation. The factors affecting image and velocity resolution are analyzed.