Vibration detection in phase-sensitive optical time domain reflectometry sensing

The distributed optical fiber vibration sensors (DOFS) have been widely explored with regards to their significant impact in sensor applications. The Phase-sensitive Optical Time Domain Reflectometry (Φ-OTDR) which is known as one of the most vigorous distributed optical fiber sensing technologies has attracted enormous research attention due to its strength in high precision measurements, fast speed response, long perimeter monitoring, as well as vibration detection abilities, among others. However, it becomes very traumatic when data of the said sensing technology meant for vibration detection is obstructed by noise. Therefore, in order to successfully enhance effective vibration detection by the Φ-OTDR sensing technology, denoising becomes very essential. The Φ-OTDR sensing data therefore was initially processed by Hilbert transform to retrieve both the real and imaginary parts of the complex signal. Then discrete wavelet transform was identified and carefully applied to obtain the desired denoised output of which phase unwrapping was performed to reveal the vibration point. In the experiment, the Signal-Noise-Ratio (SNR) of the location information was greatly improved from 16.0dB to 30.0dB on 2km sensing fiber. Therefore, the proposed method has the potential to precisely extract intrusion location from any harsh setting with strong noise background.