Research Information System
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, vol.63, no.9, pp.2730-2740, 2015 (SCI-Expanded)
Microwave imaging (MWI) systems extensively employ vector network analyzers for microwave measurements due to their high availability and accuracy. This is in contrast to theoretical models, which are naturally formulated in terms of scattered electric field vectors. Accordingly, experimental verification of MWI methods requires an intermediate step where measured scattering parameters are converted to scattered electric fields. In parallel to recent research, which formulates the Born iterative method in terms of scattering parameters, we develop formulations of two closely related qualitative inverse scattering methods-the linear sampling method and the factorization method-directly in terms of scattering parameters to avoid the intermediate conversion step. To this aim, we introduce vector S-parameters and we extend the vector Green's function for S-parameters to the dyadic case. There are certain advantages of these formulations over their electric field counterparts. First of all, the resulting formulations inherently incorporate the antenna radiation characteristics. Moreover, they reduce the measurement time since they do not require any pre- or post-measurement process. We experimentally verified the presented novel formulations against multi-frequency measurements performed inside an anechoic chamber. Obtained results show that the proposed methodologies can accurately reconstruct the shape of the targets by directly exploiting multifrequency measurements in the imaging process.