Self-Optimized Disturbance Localization of Asymmetric Fiber Interferometer Sensors Using an Enhanced Time-Delay Estimation Algorithm

Distributed disturbance detection system, grounded in fiber interferometer sensors, presents notable advantages, including extended sensing ranges and seamless integration with fiber-optic communication networks, while concurrently reducing detection complexity. Nevertheless, owing to the cumulative effect of forward light propagation, this sensing technology remains hindered by its inability to precisely localize disturbances across a vast monitoring area. To address this constraint, this article introduces and experimentally substantiates a self-optimizing disturbance localization strategy that leverages an enhanced time-delay estimation algorithm. In particular, a singular spectrum analysis technique is applied to accurately extract the time-varying distribution characteristics of sensing signals induced by disturbances. As a result, the time delay between the sensing signals can be precisely determined through a comprehensive analysis of their respective time-varying attributes. Furthermore, an algorithm rooted in the Adam gradient descent optimizer is devised to automatically calculate the crucial hyperparameters of the proposed localization strategy. Consequently, the localization accuracy of the proposed approach can be substantially improved through the utilization of the calculated optimal hyperparameter combinations. Field tests were carried out to evaluate the effectiveness of the proposed localization strategy within a distributed disturbance sensing system based on an asymmetric dual Mach-Zehnder fiber interferometer. The results indicate that, for disturbance events in a perimeter security application context, the proposed strategy attains a mean localization error of merely 6.04 m at a disturbance position of 96.84 km. Additionally, the average response time of the proposed strategy is a mere 0.348 s. Hence, it is believed that the proposed detection strategy holds significant promise in long-distance disturbance monitoring fields that rely on interferometer-based distributed fiber-optic disturbance sensors.