Distributed optical fiber sensing is an important technology for monitoring optical networks and ensuring the safety and integrity of key infrastructures.
These sensors are capable of providing position-resolved information by analyzing weak scattering reflections within silica fibers.
Common distributed fiber sensing methods include time-domain, frequency-domain, and correlation-domain techniques. Time-domain methods launch intense light pulses into the fiber and interpret backscattered signals by their time of flight to locate events along the fiber.
Frequency-domain approaches rely on coherent detection to achieve very high spatial resolution over short ranges, while correlation-domain sensing uses modulated inputs combined with delayed replicas of the signal to localize reflections effectively.
Despite their maturity, all these approaches suffer from positional dependence due to fiber attenuation, which lowers the signal-to-noise ratio at greater distances.
To address these limits, this new study examines an approach in which the sensing fiber itself forms part of a laser-like oscillation. This innovation aims to create a more uniform oscillatory sensing signal that inherently counters the degradation of sensing signals with distance.
Additionally, encoding sensing information into the spectrum allows direct access to positional data by electronically selecting the sensing location, reducing the need for complex processing.
The Study: Random Optical Parametric Oscillator Fiber Sensor
Efforts to produce laser-sensor-like distributed fiber sensing have been challenging because of the weak distributed reflectivity intrinsic to scattering and the need to localize interactions along the fiber.
Many fiber laser sensors either require a fixed mirror at the sensing location or access to both fiber ends to electronically select where sensing occurs.
More recently, electronically addressable fiber sensors (EAFS) have helped bridge the gap between point sensors and fully distributed systems by allowing a single location in the fiber to be selected electronically.
However, common EAFS methods can be constrained by practical requirements such as dual-end access or limited range under pump-depletion conditions.
In this study, a different approach was introduced involving a random optical parametric oscillator (R-OPO) that uses modulation instability (MI) and continuous weak reflections to create a laser-like sensing response without requiring a fixed mirror at the sensing location.
This configuration uses a tunable fiber Bragg grating (FBG) and two cascaded fiber segments (a “gain” fiber leading to a “sensing” fiber) so that the sensing location can be tuned electronically by changing the pump pulse repetition rate to satisfy a synchronization condition that selects a single 1 m-long section at a time.
The approach enables long-distance access (>25 km) while arbitrarily addressing 1 m-long fiber sections over a long sensing range (>1 km).
At higher repetition rates, multiple sections can simultaneously satisfy the synchronization condition, so the “single-section” operating mode depends on keeping the system within a regime where only one section is synchronized.
Under the paper’s single-section condition, however, fiber sections in the first half of the total fiber length may not be addressable. This is an explicit tradeoff of enforcing unambiguous synchronization.
Results from the R-OPO Method
The R-OPO scheme takes advantage of non-linear effects rather than avoiding them, using MI to provide distributed parametric gain that supports laser-like oscillation and strong sensing signals.
It supports both backward and forward sensing, and unlike many forward-transmission approaches, the sensing location information is readily available at both fiber ends.
A central practical feature is the proposed detection method: sensing information is encoded in the intra-pulse frequency (GHz-scale), while the pulse repetition rate provides the sensing location (kHz-scale).
In the authors’ implementation, dynamic perturbations can be recovered with only a single fast Fourier transform (FFT), supporting real-time monitoring with straightforward processing.
Performance-wise, the study reports temperature- and strain-noise-limited sensitivities of 10.73 μ°C/√Hz and 80.6 pε/√Hz, respectively.
The authors also show that four-wave-mixing (FWM) by-products inherent to R-OPO operation can be used to increase sensitivity by a factor of two compared with conventional Rayleigh-based sensors, and they propose a simple frequency-unwrapping method to extend dynamic measurement range, including continuous tracking of a ~2 °C (2.15 °C in the demonstration) temperature increase.
The paper also emphasizes practical considerations: while the system was demonstrated using a sensing fiber with enhanced reflectivity, the authors discuss paths toward operation with standard single-mode fiber, including hybrid gain mechanisms and tradeoffs between pulse width, spatial resolution, and oscillation threshold.
They note that uniform temperature changes along the long gain fiber can shift the addressed location. To manage this, they outline a measurement protocol that compensates for such drift during operation.
Towards Distributed Fibre Sensing
Laser-like, electronically addressable fiber sensors are a promising advancement in optical sensing, addressing longstanding limitations in fiber loss and signal degradation.
By integrating the sensing fiber into a laser-like R-OPO oscillation, the approach provides electronically tunable sensing locations without requiring a fixed mirror at the sensing point and enables both backward and forward measurement, with location information accessible at both fiber ends.
With >25 km of access, 1 m spatial resolution, single-FFT readout for dynamic measurements, and demonstrated sensitivities down to 10.73 μ°C/√Hz and 80.6 pε/√Hz, the R-OPO sensor offers a compelling option for monitoring vital infrastructure where only selected critical points require frequent, quantitative checks.
Journal Reference
Tovar, P. et al. Random optical parametric oscillator fiber sensor. Light Sci Appl 15, 52 (2026). DOI: 10.1038/s41377-025-02049-9,