The problem of imaging 3-D perfect electric conducting objects from scattered field measurements is addressed. Plane waves at a fixed angle of incidence and varying frequency provide the illuminating radiation whereas the scattered field is collected in far-field zone. The physical optics approximation is adopted to simplify the scattering model and the scatterers' shapes (i.e., their contour surfaces) are described as the support of delta distributions. This leads to a linear integral relationship between the scattered field and the unknown distributions, the linear distributional model, which is inverted by employing its singular value decomposition. Emphasis is placed on the study of the dyadic operator to be inverted which links the vector unknown to the scattered field vector and on the role of the polarization diversity. In particular, three different approaches are presented and compared from the computational as well as the resolution point of view. The analysis is performed numerically by means of synthetic data generated by a finite-element-method-based forward solver.
Physical Optics Imaging of 3-D PEC Objects: Vector and Multipolarized Approaches
Soldovieri F;
2010
Abstract
The problem of imaging 3-D perfect electric conducting objects from scattered field measurements is addressed. Plane waves at a fixed angle of incidence and varying frequency provide the illuminating radiation whereas the scattered field is collected in far-field zone. The physical optics approximation is adopted to simplify the scattering model and the scatterers' shapes (i.e., their contour surfaces) are described as the support of delta distributions. This leads to a linear integral relationship between the scattered field and the unknown distributions, the linear distributional model, which is inverted by employing its singular value decomposition. Emphasis is placed on the study of the dyadic operator to be inverted which links the vector unknown to the scattered field vector and on the role of the polarization diversity. In particular, three different approaches are presented and compared from the computational as well as the resolution point of view. The analysis is performed numerically by means of synthetic data generated by a finite-element-method-based forward solver.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
