V. Such

Analysis of the finite difference time domain technique to solve the Schrödinger equation for quantum devices

An extension of the finite difference time domain is applied to solve the Schrodinger equation. A systematic analysis of stability and convergence of this technique is carried out in this article. The numerical scheme used to solve the Schrodinger equation differs from the scheme found in electromagnetics. Also, the unit cell employed to model quantum devices is different from the Yee cell used by the electrical engineering community. A bound for the time step is derived to ensure stability. Several numerical experiments in quantum structures demonstrate the accuracy of a second order, comparable to the analysis of electromagnetic devices with the Yee cell.

FDTD analysis of E-sectoral horn antennas for broad-band applications

The finite-difference time-domain (FDTD) method is applied to study the performance of E-plane sectoral horn antennas designed for broad-band applications. These antennas (proposed for 6-18 GHz phased arrays) have a large bandwidth, and they are easily array integrated. These antennas have a highly complicated geometry that is modeled using a polygonal approximation in the curved boundaries. Perfect matched layers (PMLs) combined with first-order absorbing boundaries are employed to simulate the free-space environment in the FDTD mesh.

A polarizer rotator system for three-dimensional oblique incidence

A numerical model to analyze a system formed by the cascade connection of slanted strip grating plates to rotate the polarization plane of a linearly polarized wave is presented. A variant of the Generalized Scattering Matrix Theory is formulated in terms of E-type and H-type modes. These modes enable as to study three-dimensional oblique incidence upon the polarizer rotator system. A simulation program was implemented allowing the design and optimization of these systems. General criteria of design are presented, and a four-plates 60/spl deg/ polarizer rotator is designed in the 7-18 GHz band; a prototype was manufactured and tested. Comparison between the theoretical results and measurements are given, and a good agreement is found. >

Modeling of thin curved sheets with the curvilinear FDTD

The finite-difference time-domain method in general curvilinear coordinates (FDTD-GCC), or nonorthogonal FDTD, permits the analysis of arbitrary curved structures with the use of a conformal mesh. The analysis of near fields in the proximity of a thin curved dielectric sheet is a difficult task; then, ray tracing techniques and spectral numerical techniques are usually employed in the analysis of radomes. In this paper, the FDTD-GCC is modified to account for the analysis of thin curved dielectric sheets. The contravariant electric field normal to the sheet is split in two subcomponents, and new nodes are introduced where the thin sheet is located. New updating equations are inserted in the calculation of contravariant field components. The utility of the proposed technique is demonstrated in the analysis of two cylindrical radomes, the first having four not centered dipoles and the second with a centered dipole. The thickness of the radomes was 0.035-0.026 wavelengths at the central operating frequency.

T-junctions in square coaxial waveguide: a FD-TD approach

The finite difference time domain (FD-TD) method is applied to model discontinuities in square coaxial waveguide, particularly a T-junction in square coaxial cable is studied. The separation of the computation domain in a total field region and a reflected field region, and the use of a synthetic excitation constituted of a great number of monochromatic waves allow us to develop an efficient algorithm. Numerical results are given for the scattering parameters of a T-junction in the band 1-6 GHz. >

The phase center position of a microstrip horn radiating in an infinite parallel-plate waveguide

In this paper the field radiated by a flared microstrip line in an infinite parallel-plate waveguide with separation between plates of less than half a wavelength is studied. The field at any point of the parallel-plate waveguide is expressed as a summation of modal solutions of the Helmholtz equation. The phase center of the microstrip horn is calculated for uniform amplitude distributions along the horn aperture. The theory may be applied to obtain the phase and amplitude distributions along the output ports of microstrip bootlace lenses. >

Electromagnetic Scattering by a Strip Grating with Plane-Wave Three-Dimensional Oblique Incidence by Means of Decomposition into E-Type and H-Type Modes

A numerical algorithm to analyze the plane-wave three-dimensional oblique incidence on a strip grating is presented. Electromagnetic field is decomposed into vector Floquet harmonics of the E-type and H-type modes. To impose boundary conditions on the incident, reflected and transmitted waves, two integral equations of Fredholm of first kind are obtained. These equations are solved numerically with the standard Galerkin procedure, and the convergence of the algorithm is examined numerically. Since the superficial current near the edges of a conducting strip have been taken into account, the computational algorithm shows a fast convergence. Results are compared with other numerical results available in the literature to demonstrate the accuracy of the proposed method. Some numerical results are shown at normal and oblique incidence, to make clear the behaviour of the grids at different azimuth incidence angles.

Transmission properties at microwave frequencies of two-dimensional metallic lattices

The transmission properties of different metallic photonic lattices (square and rectangular) have been experimentally studied. A numerical algorithm based on time domain finite differences has been used for simulating these photonic structures. The introduction of defects in the two-dimensional metallic lattice modifies its transmission spectrum. If metal rods are eliminated from (or added to) the lattice, extremely narrow peaks are observed at some particular frequencies below (or above) the band pass edge.

Diffraction by a Rotman lens

In the paper diffraction theory is applied to obtain the field received by an aperture in the parallel-plate region of a Rotman lens when the outer arc of the lens is excited with arbitrary phase and amplitude distributions. It has been found that the receiving aperture does not detect the maximum power when it is located upon the nominal focus, it is shown that the peak power position depends on the wavelength and it moves in the radial direction. The theory has been applied to predict the power received by the focal arc apertures of a Rotman lens and satisfactory agreement between theory and experiment has been found.