Design of binary long-period fiber grating filters by the inverse-scattering method with genetic algorithm...

Area of Science:

• * Optics and Photonics
• * Materials Science and Engineering

Background:

• * Designing binary long-period fiber gratings (LPFGs) presents challenges due to nonuniform coupling strength.
• * Existing coupled-mode theory offers an approximation for modeling binary LPFGs, necessitating refinement for precise design.

Purpose of the Study:

• * To develop an optimized design approach for binary LPFG filters.
• * To enhance the accuracy of LPFG filter design by addressing limitations of coupled-mode theory.

Main Methods:

• * Utilized the Gel'fand-Levitan-Marchenko (GLM) inverse-scattering method for synthesizing grating patterns.
• * Employed a coupled-mode theory with Poisson sum formula for binary index perturbation.
• * Implemented a transfer-matrix model for analyzing nonuniform binary LPFG coupling behavior.
• * Applied a real-coded genetic algorithm for optimizing designs based on GLM synthesis and transfer-matrix analysis.

Main Results:

• * Successfully synthesized nonuniform coupling strength in binary gratings by adjusting the local duty ratio.
• * Demonstrated the effectiveness of the combined GLM and genetic algorithm approach for accurate LPFG filter design.
• * Validated the method through the design of specific filters: a flatband LPFG filter and a high-visibility all-fiber Mach-Zehnder filter.

Conclusions:

• * The presented GLM-based approach, enhanced by genetic algorithm optimization and transfer-matrix modeling, provides an effective method for designing binary LPFG filters.
• * This technique overcomes the approximations of traditional coupled-mode theory, enabling precise control over grating characteristics.
• * The successful design of advanced filter types highlights the versatility and practical applicability of the developed methodology.