Abstract
Recent burgeoning machine learning has revolutionized our ways of looking at the world. Being extraordinarily good at pattern recognition, machine learning has been widely applied to many fields to solve challenging problems. This dissertation demonstrates the applications of machine learning on scanning-free fiber-optic imaging systems (FOISs), and on the design of anti-resonant fibers. In the first part, we propose a semi-supervised learning framework called the adaptive inverse mapping (AIP) to stabilize the imaging performance through multimode fibers (MMFs). We show that if the state of the MMF is traced closely, the output images can be used as probes to correct the image reconstruction inverse mapping. Robustness is increased through the AIP method but still quite limited by the intrinsic sensitivity of the MMFs to perturbations. To further increase the robustness and the image quality of FOISs, we investigate an alternative optical fiber called glass-air Anderson localizing optical fibers (GALOFs), where randomness is intentionally introduced in the fiber cross-section. Enabled by the transverse Anderson localization (TAL), the modes in the GALOF are well confined, robust and wavelength-independent. We illustrate robust full-color cellular high-fidelity imaging through the GALOF with unsupervised learning. In the second part, I demonstrate the use of reinforcement learning in capillary structure design in hollow-core anti-resonant fibers (HC-ARFs). Moreover, inspired by the loss and dispersion spectra of the HC-ARFs, we propose a solid-core anti-resonant fiber (SC-ARF) design for fiber lasers at 2 microns. Power upscaling for fiber lasers at 2 microns requires a novel design of an all-solid active fiber that operates in the normal dispersion regime by compensating the material dispersion with the waveguide dispersion, while achieves a large mode area, low losses, single mode operations and robustness. We utilize a genetic algorithm (GA) to optimize the design parameters of the SC-ARF in terms of these multiple objectives.
Notes
If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu
Graduation Date
2022
Semester
Summer
Advisor
Schulzgen, Axel
Degree
Doctor of Philosophy (Ph.D.)
College
College of Optics and Photonics
Department
Optics and Photonics
Degree Program
Optics and Photonics
Identifier
CFE0009191; DP0026787
URL
https://purls.library.ucf.edu/go/DP0026787
Language
English
Release Date
August 2023
Length of Campus-only Access
1 year
Access Status
Doctoral Dissertation (Open Access)
STARS Citation
Hu, Xiaowen, "Machine Learning in Fiber Optics" (2022). Electronic Theses and Dissertations, 2020-2023. 1220.
https://stars.library.ucf.edu/etd2020/1220