High sensitivity torsion sensor based on Er3+ doped multicore fiber ring laser
This work reports a torsion sensor based on a fiber-ring laser (FRL), whose active element is a 58 cm segment of three inline core erbium-doped fiber (EDF-3C). In this architecture, the multicore fiber simultaneously provides optical gain and torsion transduction. The applied twist alters the inter-...
| Autores: | , , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión aceptada para publicación |
| Fecha de publicación: | 2025 |
| País: | España |
| Institución: | Universidad Pública de Navarra |
| Repositorio: | Academica-e. Repositorio Institucional de la Universidad Pública de Navarra |
| OAI Identifier: | oai:academica-e.unavarra.es:2454/55865 |
| Acceso en línea: | https://hdl.handle.net/2454/55865 |
| Access Level: | acceso abierto |
| Palabra clave: | Erbium-doped fiber ring laser Multicore fiber Torsion sensing |
| Sumario: | This work reports a torsion sensor based on a fiber-ring laser (FRL), whose active element is a 58 cm segment of three inline core erbium-doped fiber (EDF-3C). In this architecture, the multicore fiber simultaneously provides optical gain and torsion transduction. The applied twist alters the inter-core coupling, producing a gain splitting that appears as two laser peaks. The wavelength separation between the peaks encodes the torsion magnitude, whereas their relative powers reveal the twist direction. By fine-tuning the polarization controller, bidirectional operation remained stable over a ±70° range without mode extinction. The differential wavelength shift exhibited linear sensitivities of 0.013 nm/° for negative torsion and 0.019 nm/° for positive torsion. Direction discrimination is further enhanced by analyzing the peak-power difference and is additionally optimized with a function-fitting neural network trained on the measured spectra. The machine learning model achieved R² = 0.9999 and an angular detection limit of 1.28 × 10-4°. |
|---|