A spectral domain approach to modelling of EM scattering for Synthetic Aperture Radar target recognition

A Fourier-based technique for electromagnetic (EM) wave reconstruction with application to polarimetric airborne and spaceborne radar data exploitation is presented. The method is different from conventional modelling techniques for Synthetic Aperture Radar (SAR) applications as a result of the full electromagnetic treatment of field interactions with the scatterer, the possibility of introducing new and controllable feature classes for target classification, and accurate decomposition of the source impulse response function that avoids potential errors (e.g. loss of coherent information) associated with the spherical phase approximations. The capability of extracting scatterer information such as the coherent radar cross section (RCS) is explored.