Event Identification for Phase-sensitive OTDR based on Boosting Ensemble Learning

Abstract

The distributed optical fiber sensor based on phase-sensitive optical time-domain reflectometer (φ-OTDR), of which has the advantages of simple structure, low energy consumption and strong anti-electromagnetic interference capability, is widely used in dynamic sensing fields such as perimeter security. However, in practical applications, due to the extremely high sensitivity of φ-OTDR, most of the received signals are useless natural sounds caused by environmental noise and human activities, which not only limits the efficiency of the system, but also causes false alarms. In order to effectively identify and timely respond to detected disturbance signals, this paper proposed an ensemble learning algorithm for optical fiber disturbance signal detection and recognition, which combined with wavelet packet decomposition (WPD) and adaptive Boosting (Adaboost) algorithm. This paper obtained optical fiber disturbance signals of 4 kinds of intrusion/interference events through field measurement, combined with three classic performance indicators of pattern recognition to verify the performance of the proposed algorithm. Experimental results show that the total classification accuracy is high to 97.78%.