Telecoms News is part of the TechForge Publications series
TechForge
AI & Predictive Networking, Automation & Self-Optimising Networks, Broadband, Case Studies, Cloud & Edge Convergence, Digital Transformation, Enterprise & Industry Solutions, Featured News, Features, Fibreoptics & Broadband, Hot Topics, Network Infrastructure, Network Visualisation, Networks, Operator Strategies, Operators, Security, Trends, Tech & Innovation
Ryan Daws
8th June 2026
Share this story:
Tags:
Categories::
Nokia has converted a 2000km production fibre route into an active distributed sensor network using optical tomography.
Network operators manage vast stretches of fibre optic cable across terrestrial paths and seabeds, facing constant anxiety regarding infrastructure integrity. Visibility into the physical health of that fibre remains difficult, leaving administrators to generally only know whether a connection is active or broken.
High-speed fibre networks function as the invisible arteries of society, yet organisations frequently depend on stock fibres they do not explicitly own. These external segments remain a black box, masking hidden threats like fibre tapping that directly compromise cybersecurity protocols.
Invisible dangers – such as a ship dragging an anchor, construction work, or cable strain – eventually sever connections and trigger an expensive scramble to fix the fault. Reactive maintenance proves costly and archaic.
Nokia Bell Labs tackled the issue with optical network tomography, converting passive glass strands into an active distributed sensor network. The engineering team proved this concept on a live 2000-kilometre production network in collaboration with Nordic research and education network operators CSC, Sikt, and SUNET.
Sylvain Almonacil, Research Engineer at Nokia Bell Labs. said: “We are not adding new sensors. We are turning the transponder itself into a sensor to see the network from the inside.”
The architecture bypasses basic continuity tests by analysing the quality of transmitted light with high precision, deducing what happens to the cable along its journey. The core technology detects minute changes to the light’s state of polarisation. Vibrations, temperature changes, or mechanical stress alter this polarisation by forcing the orientation of the light to deviate.
The trial employs a Nokia PSC 6S silicon engine to detect these minute variations. This hardware features advanced algorithms designed to optimise network capacity, automate processes using AI, and proactively detect errors. Operating as sensitive detectors, the engines continuously monitor polarisation states to map physical stresses on the cable in near real-time without affecting data transmission speeds.
Coherent transceivers sitting at the ends of the fibre link act as the primary edge sensors, executing high-frequency measurements to collect raw polarisation data directly from the light stream. Integrating this capability directly into existing network infrastructure introduces the ability to evaluate segments spanning several different operational domains.
The edge hardware feeds raw data into centralised processing algorithms that function similarly to medical tomography. These algorithms correlate the tiny changes registered at both ends of the fibre link to pinpoint the exact location and intensity of the physical disturbance.
Optical network tomography solves the issue of securing infrastructure the operator does not physically own. Engineers convert the standard transponder into the sensor to observe the entire system from the inside, allowing the transponder to effectively see through the fibre itself.
Past operations restricted administrators to verifying only the behaviour of segments they configured themselves. The new end-to-end communication protocol ensures the route operates exactly as intended regardless of ownership.
Network administrators track specific wavelengths across various optical regions, including segments managed entirely by third-party telecom operators. Situational awareness jumps dramatically as the operator receives warnings regarding specific environmental vibrations and exact geographic locations.
The trial network employed 2000 km of SUNET’s fibre infrastructure, functioning as the real-world production environment carrying actual traffic for universities and research institutions across the Nordic region. This subjected the optical equipment to real-world environmental noise and complex variables. Impressively, the implementation required no new dedicated sensors and caused zero interference with primary customer data.
Processing the operational data over a three-week period, the Nokia Bell Labs team confirmed the digital tomography estimations perfectly matched the actual physical measurements taken from the complex multi-domain network. The trial mapped out all fibre types and exact span lengths across the entire route.
The deployment alters how telecoms infrastructure operates. Pinpointing a fibre cut traditionally takes days and requires heavy capital expenditure, but optical tomography narrows the search area from hundreds of kilometres down to a specific segment. System alerts regarding nearby digging activity allow operators to intervene before the cable suffers physical damage.
Administrators also gain deeper visibility into the network structure to identify anomalous activities like fibre tapping early, addressing these security breaches before they develop into major outages. Network components protect themselves as the transponder detects the exact moment the light orientation deviates from the AI baseline.
Nationwide fibre networks essentially become distributed acoustic and seismic sensor grids, continuously mapping physical stresses applied to the cable in near real-time. This information proves valuable regarding monitoring infrastructure like pipelines or generating early warnings regarding geological events.
Environmental data extraction happens continuously at the endpoints, feeding a central software platform that processes the firehose of information generated by continuous polarisation changes. The platform learns the normal baseline vibrations from roads, railways, and weather, allowing AI to flag anomalies representing genuine physical threats.
This all culminates in enabling a change in how operators view their physical assets, turning standard data pipelines into self-protecting sensors.
See also: Broadcom silicon bridges AI data centres and edge
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the IoT Tech Expo and Cyber Security & Cloud Expo. Click here for more information.
Telecoms is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.
Ryan Daws
Senior Editor
8th June 2026
5th June 2026
3rd June 2026
2nd June 2026
Subscribe now to get all our premium content and latest tech news delivered straight to your inbox
Operators, Research
5G-A, AI & Predictive Networking, Automation & Self-Optimising Networks, Network Infrastructure, Next-Gen RAN Architectures, Operators
5G & 6G Development, AI & Predictive Networking, Automation & Self-Optimising Networks, Featured News, Hot Topics, Market Landscape & Strategy, Network Infrastructure, Next-Gen RAN Architectures, Operator Strategies, Operators, Trends, Tech & Innovation, Voices & Analysis
5G, 6G, AI & Predictive Networking, AT&T, Automation, Cloud-Enabled Telco, Enterprise & Industry Solutions, Featured News, Features, Hot Topics, Interviews & Analysis, Market Analysis & Forecasts, Market Landscape & Strategy, Network Infrastructure, Next-Gen RAN Architectures, Operator Strategies, Operators, Private Networks & SD-WAN, Thought Leadership & Opinion, Voices & Analysis
Case Studies
3rd June 2026
5G
2nd June 2026
Security
1st June 2026
All our premium content and latest tech news delivered straight to your inbox
Telecoms Events is part of TechForge
All our premium content and latest tech news delivered straight to your inbox
Leave a Reply