CNN Classification Architecture Study for Turbulent Free-Space and Attenuated Underwater Optical OAM Communications

Authors

Patrick L. Neary, Utah State UniversityFollow
Abbie T. Watnik, U.S. Naval Research LaboratoryFollow
Kyle Peter Judd, U.S. Naval Research LaboratoryFollow
James R. Lindle, DCS CorporationFollow
Nicholas S. Flann, Utah State UniversityFollow

Document Type

Article

Journal/Book Title/Conference

Applied Sciences

Publisher

MDPI AG

Publication Date

12-8-2020

First Page

1

Last Page

19

Creative Commons License

This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

Abstract

Turbulence and attenuation are signal degrading factors that can severely hinder free-space and underwater OAM optical pattern demultiplexing. A variety of state-of-the-art convolutional neural network architectures are explored to identify which, if any, provide optimal performance under these non-ideal environmental conditions. Hyperparameter searches are performed on the architectures to ensure that near-ideal settings are used for training. Architectures are compared in various scenarios and the best performing, with their settings, are provided. We show that from the current state-of-the-art architectures, DenseNet outperforms all others when memory is not a constraint. When memory footprint is a factor, ShuffleNet is shown to performed the best.

Recommended Citation

Neary, Patrick L.; Watnik, Abbie T.; Judd, Kyle Peter; Lindle, James R.; and Flann, Nicholas S., "CNN Classification Architecture Study for Turbulent Free-Space and Attenuated Underwater Optical OAM Communications" (2020). Space Dynamics Laboratory Publications. Paper 265.
https://digitalcommons.usu.edu/sdl_pubs/265