| Summary: | A major area of recent research involves the study of and realization of double-negative (DNG) metamaterials. DNG metamaterials are bulk materials in which both the permittivity and the permeability are simultaneously negative. Theoretical studies generally assume that a DNG metamaterial has certain and specific effective parameters and studies the effects of such materials under the assumption that those effective parameters can somehow be realized. Concurrent research studies explore various inclusions, which when imbedded within a host material, may posses DNG effective parameters within a certain frequency range. These effective parameters are often obtained via various mixing formulations such as Periodic [1] or Maxwell-Garnett [2] mixing formulas which approximate the effective material parameters based on the electromagnetic properties of the single inclusion. S-Parameter extraction techniques have also been formulated [3] but operate under the assumption that the slab thickness, d, is small enough with respect to the wavelength within the slab, which can often result in performing the parameter extraction of a slab consisting of only a single inclusion in thickness. When a specific DNG metamaterial is realized and fabricated, a curve fit of the S-parameter solution is also often utilized to approximate the electric and magnetic effective properties of the material [4]. In order to bridge the gap between the theoretical studies and the metamaterial engineering, it is necessary to extract the effective parameters of realized metamaterials with as much precision as possible. In this work, we investigate the potential for applying Evolutionary Algorithms [7]. In particular, we apply Evolutionary Programming (EP) to the problem of parameter extraction of DNG metamaterials. EP has been shown to be well-suited for optimization of highly non-linear continuous parameter objective functions that may manifest epistatic behavior, and has been previously applied to a number of challenging electromagnetic problems [8, 9]. Utilizing the EP optimization, it may be possible to extract the effective parameters, including frequency dependence, from the S-Parameters of a slab of arbitrary thickness.
|
|---|
