A Method for Event Localization in Sagnac Loop Sensing System Using Sparse Signal Recovery
Jameel Ali, Ahmed Almaiman, Esam M. Almohimmah, Amr Mohamed Ragheb, Maged Abdullah Esmail, Haakon Bryhni, Saleh A. Alshebeili · Journal of Lightwave Technology · 2025
Efficient localization of events within optical fiber sensing systems is essential for many applications like pipeline monitoring and intrusion detection, where fast and accurate responses are crucial. This paper proposes a method for applying sparse signal recovery (SSR) in an optical fiber Sagnac interferometer (SI) loop-based sensing system for event detection and localization. The SSR data processing techniques have emerged to meet the demand for applications like medical imaging, wireless communication, and astronomy. This is because of the SSR's ability to (i) reconstruct signals from incomplete or noisy measurements, (ii) focus on the essential parts of the signal that make the detection more robust against noise, and (iii) process signals by acquiring fewer measurements. Here, we develop a new method for event localization in an SI loop-based sensing system using SSR. The performance of the proposed method was evaluated through simulations and experiments by characterizing (i) the statistics of the detected location; using the Average Location of the Event and Standard Deviation ($\boldsymbol{\sigma }$), and (ii) the quality of the detection system; using the Probability of False Alarm and the Probability of Detection ($\mathbf {PD}$). In the simulation, which included$\mathbf {2000}$tests for both single-event and two-event scenarios, the results indicated that (i) for a single event, the$\boldsymbol{\sigma }$is between$\mathbf {0}$and$\mathbf {\pm 50}$m when the Signal-to-Noise Ratio (SNR) exceeds$\mathbf {10}$dB and within$\mathbf {\pm 400}$m at an SNR of$\mathbf {-10}$dB, with$\boldsymbol{\sigma }$decreasing significantly as the SNR increases, eventually reaching zero; (ii) for two events,$\boldsymbol{\sigma }$is between$\mathbf {0}$and$\mathbf {\pm 100}$m when the SNR exceeds$\mathbf {10}$dB, and within$\mathbf {\pm 600}$m at an SNR of$\mathbf {-10}$dB. Experimental evaluations were conducted with five scenarios for a single event and three scenarios for two events, demonstrating a$\mathbf {PD}$of$\mathbf {100}$% in the experimental setup of$\mathbf {106.245}$km of loop length ($\mathbf {\sim\! 50}$km of effective sensing distance). Additionally, we explored the feasibility of the SSR method when integrating simultaneous sensing and data communication within the SI loop-based system by modifying the setup to support both functions.