Increasing the efficiency of single and multi-channel configurations in distributed fiber sensing

Distributed fiber sensing is a promising and fascinating field of research that can improve many existing detector systems used in oil industry, seismology, minimally invasive surgery, environmental monitoring, and even IoT. While being very compact, optical fibers can be used to implement very accurate measurements with efficient spatial resolution. This work focuses on distributed sensing applied in acoustic sensing, thermal ablation, and shape sensing. Moreover, the potential for polarization-resolved sensing is explored as well. The implemented sensing configurations are based on the interrogation scheme known as optical frequency domain reflectometry (OFDR), which collects Rayleigh backscattering signals from the fibers exposed to external stimuli. This particular interrogation enables very high resolution. It can be interfaced with optical links that consist either of single or multiple channels. However, each of the arrangements has issues that can be resolved. Single channel sensors are prone to fading, which is proposed to be corrected with the machine learning techniques. In multi-channel schemes, it is challenging to achieve efficient acquisition rates, and this work explores several multiplexing techniques that can lead to improvements. In polarization-resolved sensing, birefringence is difficult to monitor. Unspun multi-core fibers are considered as a way to solve this problem.