Dynamic Range Improvement of Backscattered Optical Signals using Signal Processing Techniques

Abstract

Optical fiber-based instruments like optical time domain reflectometry (OTDR) and distributed fiber optic sensing (DFOS) systems are widely in use over the last two decades. OTDR is critical for optical fiber testing and troubleshooting in an optical network. Besides integrity testing, one can also measure the splice losses, length of the fiber under test and the fault identification over an optical network. Similarly, the DFOS is used to detect the surrounding environmental parameters, such as temperature, strain, and vibration, etc. Both the systems as mentioned above can be designed based on the backscattering signals. However, as the signals are very weak, it poses major design challenges for designing such systems. To improve the signal to noise ratio (SNR) as well as the dynamic range of backscattering based systems, we have proposed and implemented two signal processing techniques (translation invariant wavelet thresholding (TIWT) and lifting wavelet transform-modified particle swarm optimization (LWT-MPSO)). With the TIWT signal processing technique, we have achieved a L68 dB of dynamic range improvement for Rayleigh and Brillouin backscattered signals respectively and with the LWT-MPSO signal processing technique, we have achieved a 4.03 dB of dynamic range improvement for Rayleigh and Brillouin backscattered signals respectively. For this work, a single-mode fiber (SMF) is used along with a 13 dBm of laser source power for experimenting. The signal processing techniques were implemented using MATLAB 15.0 platform. � 2020 IEEE.