A Comprehensive Survey on Machine and Deep Learning for Optical Communications

The increasing complexity of optical communication systems and networks necessitates advanced methodologies for extracting valuable insights from vast and heterogeneous datasets. Machine learning (ML) and deep learning (DL) have emerged as pivotal tools in this domain, revolutionizing data analysis...

Full description

Saved in:
Bibliographic Details
Main Authors: Mohammad Ali Amirabadi, S. Alireza Nezamalhosseini, Mohammad Hossein Kahaei, Lawrence R. Chen
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11003870/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849309567704367104
author Mohammad Ali Amirabadi
S. Alireza Nezamalhosseini
Mohammad Hossein Kahaei
Lawrence R. Chen
author_facet Mohammad Ali Amirabadi
S. Alireza Nezamalhosseini
Mohammad Hossein Kahaei
Lawrence R. Chen
author_sort Mohammad Ali Amirabadi
collection DOAJ
description The increasing complexity of optical communication systems and networks necessitates advanced methodologies for extracting valuable insights from vast and heterogeneous datasets. Machine learning (ML) and deep learning (DL) have emerged as pivotal tools in this domain, revolutionizing data analysis and enabling automated self-configuration in optical communication systems. Their adoption is fueled by the growing intricacy of systems and links, driven by numerous adjustable and interdependent parameters. This complexity is particularly evident in areas such as coherent transceivers, advanced digital signal processing, optical performance monitoring, cross-layer network optimizations, and nonlinearity compensation. While ML and DL offer immense potential, their application in optical communications is still in its early stages, with significant opportunities remaining unexplored. Many algorithms have yet to be fully deployed in practical settings, underscoring the emerging nature of this research area. This paper presents a comprehensive survey of ML and DL applications across optical fiber communication (OFC), optical wireless communication (OWC), and optical communication networking (OCN), highlighting the challenges, current advancements, and future potential of these approaches. To address the identified gaps, this survey evaluates and compares ML and DL algorithms in terms of their performance, complexity, objectives, input data, metrics, and applications in optical communication. Specific emphasis is placed on identifying how these algorithms enhance system performance and optimization. Furthermore, the advantages and limitations of existing methods are analyzed, offering a clear perspective on the role of ML and DL in this domain. The survey also includes updated visual representations and domain-specific examples to elucidate the practical applications of ML and DL in OFC, OWC, and OCN. It concludes by discussing the open challenges, proposing potential solutions, and identifying promising future research directions to foster advancements in intelligent optical communication systems.
format Article
id doaj-art-1ecdc074d5974f588e0db4475de5e5d4
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-1ecdc074d5974f588e0db4475de5e5d42025-08-20T03:54:07ZengIEEEIEEE Access2169-35362025-01-0113887948884610.1109/ACCESS.2025.356955911003870A Comprehensive Survey on Machine and Deep Learning for Optical CommunicationsMohammad Ali Amirabadi0https://orcid.org/0000-0003-4190-3380S. Alireza Nezamalhosseini1https://orcid.org/0000-0001-9274-3300Mohammad Hossein Kahaei2https://orcid.org/0000-0002-1920-659XLawrence R. Chen3https://orcid.org/0000-0002-3760-4719Department of Electrical Engineering, Ferdowsi University of Mashhad, Mashhad, IranSchool of Electrical Engineering, Iran University of Science and Technology, Tehran, IranSchool of Electrical Engineering, Iran University of Science and Technology, Tehran, IranDepartment of Electrical and Computer Engineering, McGill University, Montreal, QC, CanadaThe increasing complexity of optical communication systems and networks necessitates advanced methodologies for extracting valuable insights from vast and heterogeneous datasets. Machine learning (ML) and deep learning (DL) have emerged as pivotal tools in this domain, revolutionizing data analysis and enabling automated self-configuration in optical communication systems. Their adoption is fueled by the growing intricacy of systems and links, driven by numerous adjustable and interdependent parameters. This complexity is particularly evident in areas such as coherent transceivers, advanced digital signal processing, optical performance monitoring, cross-layer network optimizations, and nonlinearity compensation. While ML and DL offer immense potential, their application in optical communications is still in its early stages, with significant opportunities remaining unexplored. Many algorithms have yet to be fully deployed in practical settings, underscoring the emerging nature of this research area. This paper presents a comprehensive survey of ML and DL applications across optical fiber communication (OFC), optical wireless communication (OWC), and optical communication networking (OCN), highlighting the challenges, current advancements, and future potential of these approaches. To address the identified gaps, this survey evaluates and compares ML and DL algorithms in terms of their performance, complexity, objectives, input data, metrics, and applications in optical communication. Specific emphasis is placed on identifying how these algorithms enhance system performance and optimization. Furthermore, the advantages and limitations of existing methods are analyzed, offering a clear perspective on the role of ML and DL in this domain. The survey also includes updated visual representations and domain-specific examples to elucidate the practical applications of ML and DL in OFC, OWC, and OCN. It concludes by discussing the open challenges, proposing potential solutions, and identifying promising future research directions to foster advancements in intelligent optical communication systems.https://ieeexplore.ieee.org/document/11003870/Machine learningdeep learningoptical fiber communicationoptical communication networkingoptical wireless communication
spellingShingle Mohammad Ali Amirabadi
S. Alireza Nezamalhosseini
Mohammad Hossein Kahaei
Lawrence R. Chen
A Comprehensive Survey on Machine and Deep Learning for Optical Communications
IEEE Access
Machine learning
deep learning
optical fiber communication
optical communication networking
optical wireless communication
title A Comprehensive Survey on Machine and Deep Learning for Optical Communications
title_full A Comprehensive Survey on Machine and Deep Learning for Optical Communications
title_fullStr A Comprehensive Survey on Machine and Deep Learning for Optical Communications
title_full_unstemmed A Comprehensive Survey on Machine and Deep Learning for Optical Communications
title_short A Comprehensive Survey on Machine and Deep Learning for Optical Communications
title_sort comprehensive survey on machine and deep learning for optical communications
topic Machine learning
deep learning
optical fiber communication
optical communication networking
optical wireless communication
url https://ieeexplore.ieee.org/document/11003870/
work_keys_str_mv AT mohammadaliamirabadi acomprehensivesurveyonmachineanddeeplearningforopticalcommunications
AT salirezanezamalhosseini acomprehensivesurveyonmachineanddeeplearningforopticalcommunications
AT mohammadhosseinkahaei acomprehensivesurveyonmachineanddeeplearningforopticalcommunications
AT lawrencerchen acomprehensivesurveyonmachineanddeeplearningforopticalcommunications
AT mohammadaliamirabadi comprehensivesurveyonmachineanddeeplearningforopticalcommunications
AT salirezanezamalhosseini comprehensivesurveyonmachineanddeeplearningforopticalcommunications
AT mohammadhosseinkahaei comprehensivesurveyonmachineanddeeplearningforopticalcommunications
AT lawrencerchen comprehensivesurveyonmachineanddeeplearningforopticalcommunications