Synthetic-aperture radar imaging through dispersive media

In this paper we develop a method for synthetic-aperture radar (SAR) imaging through a dispersive medium. We consider the case when the sensor and scatterers are embedded in a known homogeneous dispersive material, the scene to be imaged lies on a known surface and the radar antenna flight path is an arbitrary but known smooth curve. The scattering is modeled using a linearized (Born) scalar model. We assume that the measurements are polluted with additive noise. Furthermore, we assume that we have prior knowledge about the power-spectral densities of the scene and the noise. This leads us to formulate the problem in a statistical framework. We develop a filtered-back-projection imaging algorithm in which we choose the filter according to the statistical properties of the scene and noise. We present numerical simulations for a case where the scene consists of point-like scatterers located on the ground, and demonstrate how the ability to resolve the targets depends on a quantity which we call the noise-to-target ratio. In our simulations, the dispersive material is modeled with the Fung-Ulaby equations for leafy vegetation. However, the method is also applicable to other dielectric materials where the dispersion is considered relevant in the frequency range of the transmitted signals.