Proceedings Article | 14 August 2024
KEYWORDS: Signal to noise ratio, Vibration, Background noise, Modal decomposition, Laser frequency, Denoising, Reflectometry, Mathematical optimization, Optical sensing, Sensing systems
The phase noise in phase-sensitive optical time domain reflectometer (φ-OTDR) system is primarily composed of laser frequency drift, beat frequency noise, electrical noise, and fading noise, which affect phase demodulation and vibration signal analysis. In order to enhance the signal-to-noise ratio (SNR), a phase noise suppression method based on Successive Variational Mode Decomposition (SVMD) and Pearson Correlation Coefficient (PCC) is studied and introduced. The SVMD-PCC method initially decomposes the phase signal adaptively into several intrinsic mode functions (IMFs) with different center frequencies using SVMD. Subsequently, the PCC between each IMF and the original phase signal is computed. IMFs with PCC values below a preset threshold are discarded. Finally, the remaining IMFs with PCC values exceeding the threshold are superimposed to achieve suppression of various phase noises. Compared to the VMD method, the proposed method does not require predefining the number of modes, K, thereby eliminating the need for additional optimization algorithms to determine K. This resolves the issue of performance degradation in VMD caused by inaccurate K. Furthermore, using a piezoelectric transducer (PZT) as the vibration source, experiments were designed for single-frequency vibrations at 10 Hz and 200 Hz, as well as multi-frequency vibrations at 100 Hz and 500 Hz. The results demonstrate that the SVMD-PCC method achieves a superior SNR improvement. Compared to methods such as VMD, EEMD, and EMD, the SNR improvements are 2.22 dB, 4.94 dB, and 5.00 dB, respectively, with a significantly reduced computational complexity. In summary, the SVMD-PCC method effectively suppresses various phase noises, enhances the φ-OTDR’s capability in detecting vibration events, and improves the recovery of vibration signals, thus facilitating precise distributed acoustic sensing applications with strong adaptability.
