Automatic Target Recognition for Hyperfine Resolution Synthetic Aperture Radar

The goal of this research is to understand, within the bound of recently developed fine resolution Synthetic Aperture Radar (SAR) systems, the limits of the possible of newly developed Automatic Target Recognition (ATR) concepts. SAR has developed significantly since its early inception as an invaluable remote sensing capability that stands alone compared to other sensing modalities due to its all weather, day/night and long range capabilities. In recent years SAR systems have been able to transmit and receive wider and wider bandwidth signals, in terms of the images that can be formed this produced a much finer image that is able to produce incredibly detailed imagery that has substantial potential to both detect and classify targets of interest within the scene. The resulting challenge from this is the
The area of SAR simulation has also progressed in recent years due to new capabilities in ray tracing and Finite-Difference Time-Domain (FDTD) software and more and more open source simulation tools have become available to researchers. These include RaySAR tool developed by Stefan Auger [1] and the GPRMax [2-3]. The proposed research would look to utilize open source software tools to develop a simulation environment that is able to provide high fidelity SAR imagery to feed into ATR classification techniques