Retrieval of Parameters for Layered non-Smooth Interface Media: Theory andExperiment.

[摘要] Many naturally occurring or manmade objects can be modeled as three layer mediawith non-smooth interfaces. Most of the existing forward and inverse scatteringmodels that can be applied to such media are either too inefficient or have limitedregions of validity. In this dissertation an efficient forward scattering model based onthe Extended Boundary Condition Method is developed for a three layermedium. The boundary between the first and the second layers is periodic while theboundary between the second and third layers is rough. The model is then extendedby including an arbitrarily shaped cylinder placed into the third layer. Both TM andTE polarizations and PEC and Dielectric cylinder cases are considered. The Methodof Moments (MOM) is used to obtain an impedance matrix, which is then transformedinto a T-matrix. The T-matrix is transformed into a scattering matrix andcascaded with scattering matrices for the periodic and rough interfaces to obtain ageneralized scattering matrix for the total system. A solution to the inverse problemfor a three-layer medium is developed using simulated radar data. The retrieval ofthe layered- medium parameters is accomplished by sequential nonlinear optimizaxiiition starting from the top layer and progressively characterizing the layers below.The optimization process is achieved by an efficient iterative technique built aroundthe solution of the forward scattering problem. To be efficiently utilized in the inverseproblem, the forward scattering model is simulated over a wide range of unknownsto obtain a complete set of subspace-based equivalent closed-form models that relateradar backscattering coefficients to the sought-for parameters, including the dielectricconstants of each layer and the thickness of the middle layer. The inversion algorithmis implemented as a modified conjugate-gradient-based nonlinear optimization. It isshown that this technique results in accurate retrieval of surface and subsurface parameters,even in the presence of noise. To validate forward and inverse scatteringmodels, a compact tower-based radar system is built. The data collected with theinstrument is used to demonstrate sensitivity of radar measurements to changes insoil moisture and the potential for estimating surface and subsurface parameters.

[发布机构] University of Michigan

[效力级别] Electrical Engineering

[关键词] Radar;Electrical Engineering;Engineering;Electrical Engineering