Efficient synthesis of dual-band selective filters using evolutionary methods in a 1D photonic crystal slab for near-infrared applications

摘要

In this paper, the simulated annealing (SA), genetic algorithm (GA), adaptive particle swarm optimization (APSO), and genetic swarm optimizer (GSO) algorithms are successfully applied simultaneously to the solution of optimal design of arbitrarily spaced dual-band selective filters in a one-dimensional photonic crystal (PhC-1D) based on gallium arsenide (GaAs). The synthesis of dual-band PhC-1D selective filters is obtained by acting on the Bragg grating-layer widths. The four evolutionary algorithms are employed in order to reduce the quadratic error between the desired Gaussian functions defined in advance by the user and the synthesized one to obtain a desired transmission curve. A comparative study is conducted in terms of the computed final fitness values. The results prove the superior performance of GSO over GA, SA, and APSO by synthesizing three dual-band filters operating respectively at wavelengths of (1.1, 1.25) mu m, (1.31, 1.55) mu m, and (1.75, 1.95) mu m.