Phase Unwrapping Algorithm for Distributed Acoustic Sensing Based on Unlimited Sampling and Phase Diversity

Distributed Acoustic Sensing (DAS) detects strain on optical fibers over long distances with a typical spatial resolution of a couple of meters and an acquisition rate of up to 100 kHz. DAS is utilized to diagnose events in applications like pipeline monitoring, seismic detection, structural diagnostics, and sea current observation, offering high precision for complex environments. Given the requirements of these applications, the fidelity of the data from DAS is crucial to obtain the greatest accuracy of the monitored event. However, this equipment is subject to signal distortions due to the effects of noise and the sensor’s dynamic range. To mitigate these effects, this paper introduces a new data interpretation methodology based on the unlimited sampling algorithm, which enhances the dynamic range of the sensor during the signal unwrapping process and incorporates phase diversity for noise fading reduction. This work introduces a dataset using the new algorithm compared to the state of the art. The setup follows SEAFOM guidelines using a fiber stretcher to generate signals on the fiber. An improvement of 18.44% in detecting the signal’s fundamental frequency was achieved, and a hyper-parameter mapping was conducted to enable future application in systems where the current standard DAS algorithm fails to resolve signals accurately.