Abstract: The development of few-mode fiber (FMF) has found numerous interesting applications in optical sensing. Given the remarkable capabilities of FMF, optical sensors of novel functionalities can be deployed. Here, we review the progress on state-of-the-art technology for FMF-based optical sensing. In particular, we focus on utilizing FMF for multi-parameter and absorption-based sensing. Additionally, we summarize the trials of using FMF as a compromise between single-mode fiber (SMF) and multimode fiber (MMF) to develop optical sensors of higher signal-to-noise ratio (SNR) and spatial resolution. A final discussion on the main limitations of FMF-based sensors along with prospects for further research are presented.
Introduction
In the field of fiber optics, researchers have traditionally operated between two extremes. Single-mode fibers (SMF) offer high precision and reliability but are physically limited to a single spatial channel. Conversely, multi-mode fibers (MMF) can carry significant optical power but suffer from intermodal coupling, which leads to signal scrambling and noise. Few-mode fibers (FMF) have emerged as a "Goldilocks" solution, supporting a limited number of spatial modes (typically between 2 and a few dozen). This middle ground enables novel sensing functionalities that are difficult or impossible to achieve with traditional fibers.
As illustrated in the taxonomy of FMF sensing, these fibers are primarily utilized for their unique reactions to external parameters and their role as a performance compromise between SMF and MMF. By leveraging the distinct properties of individual spatial modes, FMFs can address the long-standing challenge of parameter discrimination in optical sensing.
Methodology and the Multi-Parameter Challenge
A primary limitation of standard fiber optic sensors is the difficulty in distinguishing between different environmental stimuli. For instance, in an SMF, a change in mechanical strain often produces a signal shift nearly identical to a change in temperature. Decoupling these variables typically requires complex hardware or redundant reference fibers.
FMFs solve this through the use of a calibration matrix. Because different spatial modes (LP modes) or polarization modes within an FMF react with varying sensitivities to external stimuli, researchers can construct a system of linear equations. By measuring the simultaneous shifts of multiple modes, the signals can be mathematically "unmixed" to solve for temperature, strain, and refractive index independently.
Experimental setups, such as the one shown above, utilize polarization-maintaining few-mode fibers (PM-FMF) to characterize these responses. As seen in the linear plots, the wavelength shifts for different polarization axes (x and y) exhibit distinct slopes for both temperature and strain, providing the necessary diversity for accurate multi-parameter discrimination.
Key Insights and Performance Advantages
Modes as Independent Sensors
Each mode in an FMF can be viewed as a unique sensor. For example, certain modes may be more sensitive to bending while others are more sensitive to thermal changes. Furthermore, because different modes have different radial intensity distributions—some concentrated in the core and others extending into the cladding—FMFs can be used to reconstruct the spatial distribution of gases diffusing into the fiber cross-section.
Enhanced Signal-to-Noise Ratio (SNR)
FMFs possess a larger effective area than SMFs, allowing for the launch of higher-power pulses without triggering detrimental nonlinear effects. This results in a significantly higher SNR and extended sensing ranges for distributed sensing techniques like Raman (DTS) or Brillouin (BOTDR) scattering.
Key Results and Future Outlook
Recent studies have demonstrated the practical efficacy of FMF systems:
- Accuracy: Systems have achieved simultaneous temperature and strain sensing with accuracies of $\pm 1.7^\circ\text{C}$ and $\pm 39\,\mu\epsilon$, respectively.
- Distributed Sensing: In Raman-based temperature sensing, 2-mode FMFs have outperformed SMFs, providing a temperature resolution of $7^\circ\text{C}$ at 20 km, compared to $10^\circ\text{C}$ for SMF.
- Machine Learning Integration: The use of Convolutional Neural Networks (CNNs) for mode decomposition has achieved error rates as low as 0.5%, effectively mitigating the noise issues caused by intermodal coupling.
While intermodal coupling remains a challenge as the number of modes increases, the integration of machine learning offers a path forward. Future research is expected to focus on real-time, ML-driven decomposition, allowing FMF-based networks to scale for critical infrastructure monitoring and advanced industrial sensing.
Summary updated