Signal reconstruction and time-stability analysis in φ-OTDR with VMD-HHT

Le résumé fourni par la source

To tackle signal susceptibility to noise interference and poor time stability during multi-frequency vibrations in distributed vibration sensing with phase-sensitive optical time-domain reflectometry, this paper introduces a signal reconstruction method combining Variational Modal Decomposition (VMD) with the Hilbert-Huang Transform (VMDHHT). This method aims to address the modal aliasing issues seen in traditional Empirical Modal Decomposition (EMD) and its variants. VMD adaptively estimates the center frequencies and bandwidths of intrinsic mode functions via iterative optimization, thereby effectively reducing mode aliasing. To further improve noise robustness, the Pearson correlation coefficient is used to identify valid IMF components. Experiments utilized a 5-kilometer single-mode optical fiber as the sensing medium. Superimposed sinusoidal vibrations at 100Hz, 1000Hz, and 2000Hz were applied via piezoelectric ceramics. Phase signals resulting from these vibrations were captured using IQ demodulation technology. Compared with methods such as EMD-HHT, EEMD-HHT, and CEEMD-HHT, the proposed VMD-HHT approach shows better temporal stability. Under noise-free conditions, the relative standard deviation within 50 milliseconds significantly drops from 9.3%–120.8% in other methods to 1.1%–7.7%.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.