Project Description:
Optical fibre networks are critical to global connectivity, yet they are susceptible to accidental damage or deliberate tampering. While distributed acoustic sensing (DAS) has been widely investigated for disturbance detection, an alternative approach relies on monitoring the state of polarisation (SOP) of light propagating in the fibre. The SOP is highly sensitive to mechanical perturbations, bends, and vibrations, and therefore provides a powerful, low-intrusion method to detect physical disturbances.
By continuously tracking SOP variations, it becomes possible to identify unusual events that may signal attempts to disturb or access the fibre, such as digging, tapping, or intentional tampering. Machine learning techniques can be employed to learn the characteristic temporal and statistical patterns of SOP variations, enabling reliable detection of anomalous events.
Goal:
The goal of this project is to compare how ML-based algorithms for event detection and characterisation work for two different types of SoP sensing devices. One device has higher complexity and capable of tracking the three Stokes parameters of the receiver signal, while the second is a simple design, based on polarisation beam splitter and photo detectors. The student will first optimise the data processing of the signals from the two different sensing systems, and then compare the performance of the two.