SNR enhancement with a non-local means image-denoising method for a Φ-OTDR system

Area of Science:

• Optical Engineering
• Signal Processing
• Image Processing

Background:

• Phase-sensitive optical time-domain reflectometry (Φ-OTDR) systems utilize orthogonal pulse pairs to mitigate polarization fading.
• Polarization-maintaining optical switches (PM-PSW) introduce significant noise during optical path switching, degrading system performance.

Purpose of the Study:

• To propose and validate a novel non-local means (NLM) image-processing method for enhancing the signal-to-noise ratio (SNR) in Φ-OTDR systems.
• To address the noise limitations introduced by PM-PSW in Φ-OTDR systems.

Main Methods:

• A non-local means (NLM) algorithm is employed, leveraging the texture and self-similarity of multidimensional Rayleigh temporal-spatial images.
• The NLM algorithm estimates denoising by calculating a weighted average of pixels with similar neighborhood structures.
• Experimental validation using a Φ-OTDR system with a simulated 100 Hz vibration signal at 20.04 km and a 30 Hz PM-PSW switching frequency.

Main Results:

• The NLM method significantly improves the SNR of the vibration positioning curve.
• Before denoising, the SNR was 17.72 dB; after applying the NLM method, the SNR increased to 23.39 dB.
• The experimental results demonstrate a substantial SNR improvement of 5.67 dB.

Conclusions:

• The proposed NLM image-processing method is effective and feasible for noise reduction in Φ-OTDR systems.
• This approach offers a significant improvement in SNR compared to traditional one-dimensional signal denoising methods.
• The enhanced SNR facilitates more accurate vibration location and recovery in practical Φ-OTDR applications.