Multi-task localization based on Φ-OTDR: composite vibration recognition, synchronous localization, and co-trench position

Abstract

In optical fiber networks, ensuring reliability is crucial as both newly activated and pre-existing associated services encounter co-trenching risks and potential security threats. To address these challenges, we propose a $\Phi$-OTDR-based multi-task localization framework integrating composite vibration event recognition, synchronous localization, and co-trench position detection. Analyzing real-time vibration signals, our method achieves 95.41% event synchronous positioning, 99.50% event classification, and 92.25% co-trench location accuracy, with 98.17% robustness on 400 test samples. These results demonstrate the effectiveness of the proposed framework in enhancing the safety of optical fibers and supporting the stable operation of optical fiber networks.