Analysis and Design of Volume Photonic Metal Grid Antenna Elements

Worcester Polytechnic Institute, Worcester MA

Investigators

Abstract

0096395 Makarov A two-year program to investigate transmitting, scattering, and focusing phenomena of finite rectangular volume metal grids is proposed. The grid cell sizes are on the order of a wavelength or smaller. The project results will make a contribution to the development of large-scale antennas, including base station antennas, satellite antennas, and radars. One of the impacts could be the replacement of a reflecting metal sheet or a tiny metal texture by a specially designed 3D metal grid with frequency-selective reflectivity properties. Numerical and theoretical analyses will take up 65% of the project work while experimentations will consume the remaining 35%. Today, numerical modeling of finite metal grids (and of any finite photonic crystals) constitutes considerable difficulties and is very time consuming. The speed issue of a numerical solver is a most critical parameter for the design process: many hundreds of trials may be necessary to identify a proper metal grid structure. As part of this proposal, a new iterative solver for magnetic/electric field integral equations will be developed. Its idea is to convert the primary equation to the normal (energy) form and then apply a generalized minimum residual (GMRES) optimization. The current solver performance is reported in the project description. Preliminary studies indicate that the solver is fast and requires a few tens of iterations for complicated metal grid structures. To date the solver uses the simplest (piecewise-constant) basis functions and none of the special anti-resonant techniques. The first goal of the project is therefore a modification and further development of the solver in order to be able to compute resonant volume metal grids with 100-225 cells and with 2 x 10 to the fifth power - 1 x 10 to the sixth power boundary elements in 20-40 iterations. Next, extensive numerical modeling will be performed with respect to three grid parameters: cell spacing, thickness, and depth. The fourth parameter is grid curvature for curved meshes or a shaping factor for shaped grids. The objectives of the proposed modeling efforts are: i. Roadmaps of transmission spectra for various grid patterns. ii. Search for shaped grids providing unusual transmission/focusing properties, including higher focusing gain than equivalent solid reflectors or lenses. The first objective is becoming the standard step in the development of photonic metallic/dielectric filters. The desired pass band properties for the present project will beidentified using industrial links. The second objective is more questionable, but, as explained in the project description, may offer significant benefits. The experimental part of the work includes fabrication of the metal grids, transmission spectra measurements using a network analyzer and two small horn antennas, and surface current distribution measurements on the grid using an infrared detection system and a high-power microwave feed. The proposal includes a substantial educational component geared toward general antenna theory. The teaching plan foresees one graduate (two years) project on a fast method of moments for antenna design, one or two short undergraduate projects, and incorporation of antenna-related materials into undergraduate and graduate courses, with particular emphasis on genetic optimization algorithms. As a result of the teaching plan, an introductory inantenna toolboxla for Matlab will be created. The toolbox will include receiving and transmitting (delta gap feed) antenna elements such as wires, rectangular meshes of wires, plates, and possibly convex reflectors.