View in EDS

Deep Learning for Detection of Harmful Events in Real-World, Noisy Optical Fiber Deployments

Saved in:
Bibliographic Details
Title: Deep Learning for Detection of Harmful Events in Real-World, Noisy Optical Fiber Deployments
Authors: Sadighi, Leyla, 1989, Karlsson, Stefan, Natalino Da Silva, Carlos, 1987, Wosinska, Lena, 1951, Ruffini, Marco, Furdek Prekratic, Marija, 1985
Source: Journal of Lightwave Technology. 43(13):6092-6101
Subject Terms: fully-connected layers, One-Dimension (1D) Convolutional Neural Network (CNN), harmful vibration, nonharmful vibration, Machine Learning (ML), State of Polarization (SOP) variations, eavesdropping, Deep Learning (DL)
Description: Optical network infrastructure underpins global communication networks. It is exposed to various physical layer breaches, such as fiber cuts or eavesdropping via fiber bending, that may violate privacy or disrupt services. Analyses of State of Polarization (SOP) variations induced by external events, combined with Machine Learning (ML) techniques, can contribute to early identification of events and categorization of potential threats. However, real-world deployment of automated threat detection and mitigation faces many challenges, including the inconsistencies between controlled laboratory settings, often used for dataset collection for ML training, and real-world, noisy environments. In this paper, we study the detection of external disturbances in real-world fiber installations by analyzing the induced changes in the SOP of optical signals. We develop a suite of Deep Learning (DL) models, including One Dimension (1D) Convolutional Neural Network (CNN) and fullyconnected dense layers, for the detection of harmful events in noisy environments comprising a shorter and a longer fiber link installation with overlapping external disturbances. The proposed approach employs an optical analyzer to capture SOP changes resulting from mechanical or acoustic vibrations, as well as eavesdropping attempts. Upon careful tuning of the DL models' hyper-parameters, 98.57% accuracy is obtained for the shorter, and 92.26% for the longer link installation.
File Description: electronic
Access URL: https://research.chalmers.se/publication/545956
https://research.chalmers.se/publication/545956/file/545956_Fulltext.pdf
Database: SwePub
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://research.chalmers.se/publication/545956#
    Name: EDS - SwePub (s4221598)
    Category: fullText
    Text: View record in SwePub
  – Url: https://resolver.ebscohost.com/openurl?sid=EBSCO:edsswe&genre=article&issn=07338724&ISBN=&volume=43&issue=13&date=20250101&spage=6092&pages=6092-6101&title=Journal of Lightwave Technology&atitle=Deep%20Learning%20for%20Detection%20of%20Harmful%20Events%20in%20Real-World%2C%20Noisy%20Optical%20Fiber%20Deployments&aulast=Sadighi%2C%20Leyla&id=DOI:10.1109/JLT.2025.3557748
    Name: Full Text Finder
    Category: fullText
    Text: Full Text Finder
    Icon: https://imageserver.ebscohost.com/branding/images/FTF.gif
    MouseOverText: Full Text Finder
  – Url: https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=EBSCO&SrcAuth=EBSCO&DestApp=WOS&ServiceName=TransferToWoS&DestLinkType=GeneralSearchSummary&Func=Links&author=Sadighi%20L
    Name: ISI
    Category: fullText
    Text: Nájsť tento článok vo Web of Science
    Icon: https://imagesrvr.epnet.com/ls/20docs.gif
    MouseOverText: Nájsť tento článok vo Web of Science
Header DbId: edsswe
DbLabel: SwePub
An: edsswe.oai.research.chalmers.se.e4e387e3.e138.435f.bff6.57f31b64b1a5
RelevancyScore: 1115
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 1114.736328125
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Deep Learning for Detection of Harmful Events in Real-World, Noisy Optical Fiber Deployments
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Sadighi%2C+Leyla%22">Sadighi, Leyla</searchLink>, 1989<br /><searchLink fieldCode="AR" term="%22Karlsson%2C+Stefan%22">Karlsson, Stefan</searchLink><br /><searchLink fieldCode="AR" term="%22Natalino+Da+Silva%2C+Carlos%22">Natalino Da Silva, Carlos</searchLink>, 1987<br /><searchLink fieldCode="AR" term="%22Wosinska%2C+Lena%22">Wosinska, Lena</searchLink>, 1951<br /><searchLink fieldCode="AR" term="%22Ruffini%2C+Marco%22">Ruffini, Marco</searchLink><br /><searchLink fieldCode="AR" term="%22Furdek+Prekratic%2C+Marija%22">Furdek Prekratic, Marija</searchLink>, 1985
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <i>Journal of Lightwave Technology</i>. 43(13):6092-6101
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22fully-connected+layers%22">fully-connected layers</searchLink><br /><searchLink fieldCode="DE" term="%22One-Dimension+%281D%29+Convolutional+Neural+Network+%28CNN%29%22">One-Dimension (1D) Convolutional Neural Network (CNN)</searchLink><br /><searchLink fieldCode="DE" term="%22harmful+vibration%22">harmful vibration</searchLink><br /><searchLink fieldCode="DE" term="%22nonharmful+vibration%22">nonharmful vibration</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+Learning+%28ML%29%22">Machine Learning (ML)</searchLink><br /><searchLink fieldCode="DE" term="%22State+of+Polarization+%28SOP%29+variations%22">State of Polarization (SOP) variations</searchLink><br /><searchLink fieldCode="DE" term="%22eavesdropping%22">eavesdropping</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+Learning+%28DL%29%22">Deep Learning (DL)</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: Optical network infrastructure underpins global communication networks. It is exposed to various physical layer breaches, such as fiber cuts or eavesdropping via fiber bending, that may violate privacy or disrupt services. Analyses of State of Polarization (SOP) variations induced by external events, combined with Machine Learning (ML) techniques, can contribute to early identification of events and categorization of potential threats. However, real-world deployment of automated threat detection and mitigation faces many challenges, including the inconsistencies between controlled laboratory settings, often used for dataset collection for ML training, and real-world, noisy environments. In this paper, we study the detection of external disturbances in real-world fiber installations by analyzing the induced changes in the SOP of optical signals. We develop a suite of Deep Learning (DL) models, including One Dimension (1D) Convolutional Neural Network (CNN) and fullyconnected dense layers, for the detection of harmful events in noisy environments comprising a shorter and a longer fiber link installation with overlapping external disturbances. The proposed approach employs an optical analyzer to capture SOP changes resulting from mechanical or acoustic vibrations, as well as eavesdropping attempts. Upon careful tuning of the DL models' hyper-parameters, 98.57% accuracy is obtained for the shorter, and 92.26% for the longer link installation.
– Name: Format
  Label: File Description
  Group: SrcInfo
  Data: electronic
– Name: URL
  Label: Access URL
  Group: URL
  Data: <link linkTarget="URL" linkTerm="https://research.chalmers.se/publication/545956" linkWindow="_blank">https://research.chalmers.se/publication/545956</link><br /><link linkTarget="URL" linkTerm="https://research.chalmers.se/publication/545956/file/545956_Fulltext.pdf" linkWindow="_blank">https://research.chalmers.se/publication/545956/file/545956_Fulltext.pdf</link>
PLink https://erproxy.cvtisr.sk/sfx/access?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsswe&AN=edsswe.oai.research.chalmers.se.e4e387e3.e138.435f.bff6.57f31b64b1a5
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1109/JLT.2025.3557748
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 10
        StartPage: 6092
    Subjects:
      – SubjectFull: fully-connected layers
        Type: general
      – SubjectFull: One-Dimension (1D) Convolutional Neural Network (CNN)
        Type: general
      – SubjectFull: harmful vibration
        Type: general
      – SubjectFull: nonharmful vibration
        Type: general
      – SubjectFull: Machine Learning (ML)
        Type: general
      – SubjectFull: State of Polarization (SOP) variations
        Type: general
      – SubjectFull: eavesdropping
        Type: general
      – SubjectFull: Deep Learning (DL)
        Type: general
    Titles:
      – TitleFull: Deep Learning for Detection of Harmful Events in Real-World, Noisy Optical Fiber Deployments
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Sadighi, Leyla
      – PersonEntity:
          Name:
            NameFull: Karlsson, Stefan
      – PersonEntity:
          Name:
            NameFull: Natalino Da Silva, Carlos
      – PersonEntity:
          Name:
            NameFull: Wosinska, Lena
      – PersonEntity:
          Name:
            NameFull: Ruffini, Marco
      – PersonEntity:
          Name:
            NameFull: Furdek Prekratic, Marija
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 07338724
            – Type: issn-print
              Value: 15582213
            – Type: issn-locals
              Value: SWEPUB_FREE
            – Type: issn-locals
              Value: CTH_SWEPUB
          Numbering:
            – Type: volume
              Value: 43
            – Type: issue
              Value: 13
          Titles:
            – TitleFull: Journal of Lightwave Technology
              Type: main
ResultId 1