A dyadic target model for multistatic SAR/ISAR imaging

Current SAR/ISAR imaging algorithms rely upon the assumption that the area under observation consists of a superposition of infinitesimally small isotropic scatterers (i.e., the point scatterer model). This approximation fails to capture the real-world scattering mechanisms occurring within the targets under illumination. This paper proposes an imaging technique based upon the assumption that targets may be modeled as a superposition of infinitesimally small dipoles. The orientation of each dipole is encapsulated in a dyadic contrast function. The image reconstruction, i.e., retrieval of the dyadic reflectivity function from measured data, will provide information describing the shape and the direction of predominant edges of the target.