Distributed fibre sensing turned existing cable into an instrument
Backscatter from an ordinary telecom fibre reveals strain and temperature along its whole length. The sensing infrastructure was already buried.
Send a pulse down an optical fibre and a small fraction of the light scatters back from every point along it. Analysing that backscatter as a function of return time gives a measurement distributed along the fibre — every metre becoming a sensor, over tens of kilometres.
The three mechanisms
Rayleigh scattering responds to strain and vibration. Interferometric analysis of phase changes yields distributed acoustic sensing, sensitive enough to detect footsteps near a buried cable.
Brillouin scattering shifts in frequency with strain and temperature, giving absolute measurements suited to structural monitoring over long spans.
Raman scattering has a temperature-dependent component, giving distributed temperature sensing widely used in fire detection and power cable monitoring.
Why deployment accelerated
The economics changed when it became clear that dark fibre in existing telecom cable works as a sensor. Utilities, railways, pipeline operators and telecom companies own enormous quantities of installed fibre, and much of it is unused.
Turning that into a sensing network requires an interrogator at one end. No trenching, no new cable, no sensor installation.
Where it is being used
Pipeline monitoring for leaks and third-party interference, where a machine digging near a buried line produces a distinctive signature.
Rail for train location, track condition and rockfall detection.
Perimeter security over long boundaries.
Seismology, where telecom cables — including submarine ones — have been used as dense seismic arrays in regions with few instruments, which has produced genuinely new scientific data.
Well monitoring in oil, gas and geothermal, where fibre in a borehole reports flow and temperature along its depth.
The constraint
Data volume and interpretation. An interrogator generates continuous measurements from thousands of channels, and the raw signal is not the answer — distinguishing an excavator from a passing truck requires classification trained on real events.
That has made signal processing and machine learning the competitive ground rather than the optics, which are relatively standardised. It is also why deployments succeed or fail on the quality of their event libraries rather than on interrogator specifications.
This article was produced by the LasersNews AI desk and reviewed by our editors.
Related reading
Integrated photonics narrows the gap between laboratory result and manufacturable part
Foundry access and standardised process design kits are doing for photonics roughly what they did for electro…
Metasurface optics are running into the manufacturing question
Flat lenses built from sub-wavelength structures work. Producing them at volume, at yield, with the tolerance…
Photonics found its foundry model, and the effect resembles electronics in the 1980s
Shared process design kits and multi-project wafer runs let a small team design a photonic chip without ownin…