2013/02/19 by Gregory E. Newstadt, Newstadt, Gregory E., Edmund G. Zelnio +3
Engineering · #Advanced SAR Imaging Techniques #FOS: Computer and information sciences #Information Theory (cs.IT) #Sparse and Compressive Sensing Techniques #Synthetic Aperture Radar (SAR) Applications and Techniques
paper · pdf · doi:10.48550/arxiv.1302.4680
openalex publication_date 2013/02/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In synthetic aperture radar (SAR), images are formed by focusing the response\nof stationary objects to a single spatial location. On the other hand, moving\ntargets cause phase errors in the standard formation of SAR images that cause\ndisplacement and defocusing effects. SAR imagery also contains significant\nsources of non-stationary spatially-varying noises, including antenna gain\ndiscrepancies, angular scintillation (glints) and complex speckle. In order to\naccount for this intricate phenomenology, this work combines the knowledge of\nthe physical, kinematic, and statistical properties of SAR imaging into a\nsingle unified Bayesian structure that simultaneously (a) estimates the\nnuisance parameters such as clutter distributions and antenna miscalibrations\nand (b) estimates the target signature required for detection/inference of the\ntarget state. Moreover, we provide a Monte Carlo estimate of the posterior\ndistribution for the target state and nuisance parameters that infers the\nparameters of the model directly from the data, largely eliminating tuning of\nalgorithm parameters. We demonstrate that our algorithm competes at least as\nwell on a synthetic dataset as state-of-the-art algorithms for estimating\nsparse signals. Finally, performance analysis on a measured dataset\ndemonstrates that the proposed algorithm is robust at detecting/estimating\ntargets over a wide area and performs at least as well as popular algorithms\nfor SAR moving target detection.\n