What fiber optic sensing means for structural health monitoring

Fiber optic sensing for structural health monitoring uses optical fibers — the same glass threads that carry internet traffic — as precision measurement instruments embedded in or bonded to civil structures. Rather than transmitting data between two points, the fiber itself becomes the sensor: changes in strain, temperature, or vibration along its length alter the optical signal in measurable ways, giving engineers a continuous, real-time picture of structural condition without the corrosion, electromagnetic interference, or long-term drift that afflict conventional electrical gauges.

The appeal for large civil infrastructure — bridges, dams, tunnels, retaining walls, pipelines — is straightforward. These structures are enormous, often operate in harsh environments, and must remain safe for decades. Conventional electrical strain gauges or thermocouples can only tell you what is happening at the exact point where each instrument is installed. Fiber optic systems, depending on the interrogation method, can monitor discrete critical points or every centimetre along a fiber run that stretches for tens of kilometres. The choice between those two paradigms — single-point versus distributed — is the central technical and economic decision in any fiber optic structural monitoring deployment, and it is one that organisations such as IEEE and ASCE have extensively documented in the context of infrastructure resilience.

Understanding the distinction begins with the physics. All fiber optic sensing exploits the interaction between light and the glass medium through which it travels. Strain compresses or stretches the fiber, changing the optical path length and altering the light’s properties. Temperature causes thermal expansion and changes the refractive index. The two dominant sensing paradigms — single-point (most commonly Fiber Bragg Grating) and distributed (Brillouin or Rayleigh scattering-based) — exploit these optical effects in fundamentally different ways, with profoundly different implications for where they are useful and how much they cost to deploy.

Single-point sensing: how Fiber Bragg Gratings work and where they excel

A Fiber Bragg Grating (FBG) sensor is a periodic variation of the refractive index inscribed directly into the core of an optical fiber using ultraviolet laser exposure. This grating acts as a wavelength-selective mirror: when broadband light travels through the fiber, the grating reflects a very narrow band of wavelengths — the Bragg wavelength — while transmitting all others. When the fiber is strained or its temperature changes, the grating period and the effective refractive index both shift, moving the reflected Bragg wavelength in a predictable, linear fashion. An interrogation unit measures this wavelength shift — typically with a resolution of around 1 picometer — and converts it directly into a strain or temperature value at that precise location.

The practical strength of FBG sensors lies in their multiplexing capability. Because each grating can be written to reflect a different wavelength, many FBGs — typically 8 to 80 per fiber channel, depending on the interrogator — can be read simultaneously over a single fiber run. This means a single cable routed through a bridge deck can carry dozens of independent measurement points, each reporting independently. FBG interrogators are also fast: high-speed systems can acquire data at kilohertz sampling rates, making FBGs well suited to dynamic load monitoring, vibration analysis, and impact detection — applications where millisecond-resolution data matters.

The constraint is spatial. Each FBG is inscribed at a fixed location during manufacturing. If a crack initiates between two gratings — in a section of the structure that was not instrumented — the FBG array will not detect it. For well-characterised structures where the critical measurement points are known in advance (specific bearings, post-tensioned tendon anchorages, welded joints, pile heads), this is a manageable limitation. For structures where damage location is unpredictable, it becomes a fundamental gap.

FBG sensors are also inherently passive — they require no power at the sensor itself — and are highly stable over long periods. These properties, combined with their immunity to electromagnetic interference, have made them the dominant choice for instrumented laboratory specimens, long-span bridge monitoring programmes, and aerospace structural testing, as documented in standards developed by bodies including ISO.

Distributed sensing: Brillouin and Rayleigh scattering across kilometres of infrastructure

Distributed fiber optic sensing treats the entire length of an optical fiber as a continuous array of measurement points, rather than a conduit between discrete sensors. Every centimetre of fiber is simultaneously a sensor, an antenna, and a transmission medium. This is made possible by the natural scattering of light within the glass: as photons travel through the fiber, a tiny fraction are scattered back towards the source by interactions with the glass material itself. The spectral and temporal characteristics of this backscattered light encode information about the local physical conditions — strain and temperature — at every point along the fiber’s length.

Brillouin scattering: long-range strain and temperature profiling

Brillouin scattering occurs when photons interact with thermally excited acoustic waves (phonons) in the glass. The frequency of the backscattered light is shifted relative to the input — the Brillouin frequency shift — and this shift changes linearly with both strain and temperature. By measuring the Brillouin frequency shift as a function of position along the fiber, an interrogation system can produce a complete strain and temperature profile of the entire fiber run. Two principal interrogation architectures are used in civil engineering: Brillouin Optical Time Domain Reflectometry (BOTDR), which uses a single-end pulsed measurement and is well suited to long inaccessible structures; and Brillouin Optical Time Domain Analysis (BOTDA), which uses two-end stimulated Brillouin scattering for improved signal-to-noise ratio and spatial resolution.

“Brillouin-based distributed sensing can profile strain and temperature continuously over distances of up to approximately 30 kilometres — covering an entire motorway bridge approach, a dam abutment, or a long tunnel section from a single interrogation unit at one end.”

The spatial resolution of Brillouin systems — typically around 1 metre — is sufficient for detecting localised settlement, crack zones, or temperature anomalies in large civil structures. However, a fundamental challenge is that the Brillouin frequency shift responds to both strain and temperature simultaneously. Separating the two contributions requires either a reference fiber (mechanically decoupled from the structure to measure temperature only) or a dual-parameter measurement technique. This adds complexity and cost to deployments where both quantities are independently important.

Rayleigh scattering: high-resolution sensing over shorter distances

Rayleigh scattering arises from random, nanometre-scale density fluctuations frozen into the glass during fiber manufacture. Each fiber has a unique, stable Rayleigh backscatter pattern — effectively a physical fingerprint. When the fiber is strained or its temperature changes, this pattern shifts in a way that can be detected by Optical Frequency Domain Reflectometry (OFDR). Because OFDR compares the current backscatter pattern against a stored baseline, it achieves very high spatial resolution — down to sub-millimetre scales — enabling detailed mapping of strain gradients across short structural elements such as composite panels, short tunnel linings, or laboratory specimens.

The trade-off is sensing range. Rayleigh OFDR systems are typically limited to a few hundred metres, compared with the tens of kilometres achievable with Brillouin systems. For large civil infrastructure such as long-span bridges or major dam faces, Rayleigh sensing is therefore more commonly used for detailed characterisation of specific zones rather than whole-structure monitoring. Research published through bodies such as SPIE has documented its use in composite bridge deck sections and short tunnel segments where millimetre-resolution strain mapping is required.

Comparing the two approaches: spatial coverage, resolution, and cost trade-offs

The most important practical distinction between single-point FBG sensing and distributed sensing is not technical performance — it is the fundamental difference in what each system assumes about where damage will occur. FBG arrays assume that the engineer knows, in advance, which locations on a structure are most critical and most likely to experience significant strain or temperature changes. Distributed sensing makes no such assumption: it monitors everything, everywhere, continuously.

In terms of spatial resolution, FBG sensors provide essentially point measurements with sub-microstrain precision at their exact location, while Brillouin distributed systems typically achieve spatial resolutions of around 0.5 to 1 metre over long ranges. Rayleigh OFDR systems bridge the gap with sub-millimetre resolution, but only over short distances. For detecting a 5-centimetre crack zone in a 500-metre tunnel lining, a Brillouin system with 1-metre resolution may be adequate; for mapping the strain gradient across a 20-centimetre composite joint, only Rayleigh or FBG can provide the necessary detail.

Sampling speed is another differentiator. FBG interrogators can operate at kilohertz rates, making them suitable for dynamic structural monitoring — capturing the response of a bridge deck to a passing vehicle, for example, or detecting impact events in real time. Distributed sensing systems, particularly Brillouin-based ones, are fundamentally slower: a single measurement sweep of a long fiber can take seconds to minutes, depending on the required spatial resolution and signal averaging. This makes distributed sensing better suited to quasi-static monitoring — detecting slow settlement, long-term creep, or gradual temperature changes — than to high-frequency dynamic events.

Cost considerations are multi-dimensional. Individual FBG sensors are inexpensive, but the interrogation hardware (tunable laser or broadband source with spectrum analyser) adds cost, and the labour involved in precisely positioning and bonding many individual gratings to a large structure is substantial. Distributed sensing requires only a single continuous fiber (relatively cheap to install) but the interrogation hardware — particularly for Brillouin systems — is significantly more expensive than a comparable FBG readout unit. For very long structures where spatial coverage is the priority, the economics of distributed sensing improve markedly; for short structures with a small number of critical measurement points, FBG arrays are typically more cost-effective.

Choosing the right technology for bridges, dams, and tunnels

The choice between single-point and distributed fiber optic sensing is ultimately driven by the specific failure modes, geometry, and monitoring objectives of each structure type. For each major category of large civil infrastructure, the two paradigms serve different — and often complementary — roles.

Bridges

Long-span cable-stayed and suspension bridges present a strong case for FBG sensing at known critical locations: stay cable anchorages, deck expansion joints, bearing pads, and the connection zones between deck sections and pylons are all points where strain concentrations are predictable and where high-speed dynamic data is valuable. FBG arrays installed at these points during construction provide a permanent, high-resolution record of structural behaviour under traffic and wind loading. Distributed sensing — typically Brillouin — is used in parallel to monitor the overall strain profile of long deck sections, detect any settlement or differential movement between piers, and provide a safety net for damage that might initiate away from the instrumented points. According to guidance from infrastructure bodies including FHWA, integrated monitoring systems combining both approaches are increasingly specified for major bridge programmes.

Dams

Dam safety monitoring is an application where distributed sensing has a particularly compelling advantage. A large embankment or concrete gravity dam may be hundreds of metres long and tens of metres tall. The failure modes of concern — internal erosion (piping), seepage, differential settlement, and cracking — can initiate at any point within the body of the dam or along its foundation contact. Distributed Brillouin sensing fibers embedded in the dam body or along drainage galleries can detect anomalous temperature gradients (indicating unexpected seepage pathways) or strain concentrations (indicating incipient cracking or settlement) anywhere along their length, without requiring the engineer to have predicted the exact failure location. FBG sensors are used at specific instrumented sections — crest monitoring points, gallery instruments, and foundation anchors — where high-precision point data is required alongside the distributed profile.

Tunnels

Tunnel linings present a geometry particularly well suited to distributed sensing: a long, cylindrical structure where damage (cracking, spalling, invert heave, convergence) can occur at any point along the alignment. Distributed Brillouin fibers bonded to or cast into the lining provide continuous strain monitoring along the full tunnel length. Where detailed characterisation of a specific zone — a known geological transition, a repair section, or an area of previous distress — is required, Rayleigh OFDR sensing can provide the millimetre-resolution strain mapping needed to understand the local structural behaviour. The combination of the two technologies, with the distributed system providing the broad spatial coverage and the high-resolution system providing detailed local characterisation, represents current best practice for major tunnel monitoring programmes, as reflected in technical guidance from bodies such as WIPO‘s patent database, which documents extensive filings in tunnel structural health monitoring instrumentation.