Near-field high-resolution imaging based on inverse equivalent source methods

Standard radar imaging methods work commonly with time-domain or spectral domain back-propagation techniques, which are restricted in terms of flexibility and spatial resolution. In this work, we investigate inverse source methods with respect to their imaging capabilities. In an inverse source method, we explicitly solve a linear inverse problem for the given, possibly irregularly distributed measurement samples, and with full probe correction. Moreover, we assume certain a priori knowledge about the location of the equivalent sources on the device under test (DUT). Such a formulation can be set up in form of a radiation problem or in form of a scattering problem, where the sources are rather scattering coefficients than current distributions. Our basic formulation works with time-harmonic high-frequency excitation where, however, several frequencies can be processed in order to combine the results to images with a larger information content.