Projecte assignat
Títol: Monitoring of optical fiber networks through the analysis of the state of polarization in quasi-coherent receivers
Director/a: LAZARO VILLA, JOSE ANTONIO
Departament: TSC
Títol: Monitoring of optical fiber networks through the analysis of the state of polarization in quasi-coherent receivers
Data inici oferta: 22-06-2026 Data finalització oferta: 22-02-2027
Estudis d'assignació del projecte:
GR ENG SIS TELECOMUN
| Tipus: Individual | |
| Lloc de realització: |
|
| Paraules clau: | |
| Optical network monitoring, State of polarization, Quasi-coherent receivers, Digital signal processing, Machine learning, Fiber optic sensing, Vibration detection, Spectral analyzers | |
| Descripció del contingut i pla d'activitats: | |
| This Bachelor's Thesis addresses the repurposing of quasi-coherent receivers as spectral analyzers for optical network status monitoring through State of Polarization (SoP) measurements. The main objective is to evaluate the capability of these receivers to detect external mechanical perturbations based on variations in the received optical signal parameters.
Once acquired, the signal will be processed using specific software tools to extract relevant time and frequency domain parameters. This includes studying variations in SoP, received power, spectral characteristics, and other indicators correlated with the perturbations applied to the fiber. The joint analysis of these parameters will allow the identification of patterns associated with different types of vibrations and the evaluation of their impact on the optical signal. In addition to individual parameter analysis, potential correlations between SoP variations and other optical signal indicators will be studied. Statistical analysis and digital signal processing techniques will be employed to identify which parameters exhibit higher sensitivity to mechanical perturbations. This study will evaluate whether combining multiple variables improves detection capability compared to using SoP exclusively, and determine the most suitable feature set for machine learning algorithms. Furthermore, the option of extracting additional information on the nature of the perturbations from time-domain and spectral parameters to complement SoP data will be explored. The experimental phase will be carried out in a controlled laboratory environment using a vibration generator to induce mechanical excitations of known frequency and amplitude on a supporting plate. The optical fiber will be arranged in various geometric configurations to study how factors like exposed length, fiber layout, or vibration application points affect system sensitivity. The system's response to different frequency and amplitude ranges will also be analyzed to characterize detection limits and achievable resolution. An accelerometer placed near the fiber will provide synchronized acceleration measurements to serve as a physical reference for validating the results. Data from both the optical receiver and the accelerometer will be stored and processed using libraries like Pandas, NumPy, and SciPy to perform cleaning, time synchronization, filtering, noise reduction, and sample normalization. Subsequently, a labeled dataset will be generated to analyze the relationship between mechanical perturbations and optical signal variations. Machine learning and deep learning techniques will be evaluated on this dataset to develop models capable of detecting, classifying, and estimating vibration intensity. Considered methods include supervised classification, anomaly detection, and neural networks, selecting those that offer the best trade-off between accuracy, computational complexity, and generalization. Finally, the feasibility of utilizing quasi-coherent receivers as a low-cost solution for monitoring and sensing in fiber networks will be analyzed. This includes exploring applications in early warning systems, critical infrastructure supervision, failure prevention, and future distributed sensing applications based on parameters extracted directly from communications receivers. |
|
| Orientació a l'estudiant: | |
| GR ENG SIS TELECOMUN background.
Schedule flexible. |
|
| Requereix activitats hardware: Si | |
| Requereix activitats software: Sí Sistema operatiu: Linux Disc (Gb): 250 | |
| Horari d'atenció a estudiants per a l'assignació de projecte: Dimarts 14:00 |
|