| The use of high-frequency asymptotic techniques, such as the Uniform Theory of Diffraction (UTD) and the Method of Equivalent Currents (MEC), is considered mature for predicting the electromagnetic scattering in principal planes for perfectly conducting geometries. For more complex configurations involving oblique angles of incidence or imperfectly conducting materials, either gaps in the theory or in the methods of applying the theory exist. Many interesting problems exist which could benefit from the application of high-frequency techniques to their solution. These range from the design of low-observable vehicles to the analysis of wireless communications systems. Although advances in computer technology resulting in faster and larger computational resources have enabled the use of more exact numerical techniques, such as the Moment Method (MM), the Finite Element Method (FEM), and the Finite-Difference Time-Domain (FDTD) technique, for many modern problems in electromagnetics, the complexity of many problems will always tend to exceed the computational resources available. It is, therefore, important to continue to develop and to expand the capabilities of high-frequency asymptotic techniques. In this dissertation, three canonical configurations, each of which exhibits a unique characteristic or mechanism which present high-frequency models have not adequately addressed, are examined and modeled using the UTD and the MEC.; The principal-plane scattering from a coated conducting plate, with a lossy coating of finite thickness, is studied first. Two different UTD models for this geometry are examined and results are compared to physical optics (PO) and measured data. There are several mechanisms of interest which can be isolated and analyzed using this structure. The mechanisms studied in this dissertation are higher-order diffraction and surface-wave interactions; diffraction from a perfectly conducting surface coated with a finite-thickness, lossy coating; and grazing and near-grazing diffraction.; Next, the coated conducting dihedral corner reflector is modeled using a PO model and a UTD model. Results are compared to experimental data. This is an important structure for studying reflection from coated conducting surfaces because reflection terms are the primary contributors to the overall scattered field. In this dissertation, two different reflection coefficients, which account for the effect of the lossy coating, are examined. In addition, the PO and UTD models are compared to illustrate the necessity of modeling diffraction mechanisms for this geometry at angles away from the specular. Although reflection terms dominate the scattered field, a complete model must also account for diffraction.; The UTD cannot easily and adequately model nonprincipal-plane scattering. The MEC appears to be the most promising high-frequency technique for doing this. The final problem considered in this dissertation is nonprincipal-plane scattering from a perfectly conducting rectangular plate. In order to adequately model plate scattering, it is necessary to account for corner scattering. A method of incorporating an MEC based corner scattering formulation is illustrated for the perfectly conducting plate. Heuristic solutions for improving this model are also discussed. Theoretical results are compared to experimental data. |
