In remote sensing applications, multilayered and multiregion rough surfaces are encountered. Such applications include the estimation of sea ice depth, the detection of oil spills on oceans, and the characterization of scattering from conglomerates of ice, water and soil.

The numerical solution of scattering from such multilayered multiregion rough surfaces requires fast and memory efficient computational techniques. The Steepest Descent Fast Multipole Method (SDFMM) is an integral equation based multilevel algorith for efficientlty analyzing scattering from two dimensional rough surfaces. In the following Figure, a multilayered rough surface is shown with different roughness parameters for the top and the bottom surfaces. These parameters are the root mean square height (0.03 lambda for the top surface and 0.02 lambda for the bottom surface) and the correlation length (0.5 lambda for the top surface and 0.6 lambda for the bottom surface). The relative permittivity of the layer is 2.0.

Geometry of the rough surface

The SDFMM allows the rapid and efficient simulation of scattering from such complex surfaces. The following Figure shows the bistatic RCS of the rough surface from the previous Figure, with an wave incident at 20 degrees. in terms of the scattering angle for both the PEC bottom and the dielectric bottom.

Bistatic RCS for both PE and dielectric bottoms

The above work is a collaboration between Dr. Vikram Jandhyala, Dr. Magda El-Shenawee, and Prof. Eric Michielssen. Please send suggestions, comments, and inquiries to: magda@decpc1.ece.uiuc.edu.