Performance enhancement of distributed optical fibre sensors

One of the most significant challenges of Brillouin optical time domain analysis (BOTDA) sensing is to conduct accurate measurements of temperature or strain in the shortest possible time. Several novel techniques are introduced in this thesis to reduce the measurement time and increase the accuracy of BOTDA sensors.;Investigations in BOTDA sensors show that ensemble averaging, the conventional method of denoising signals in BOTDA sensors, is not a time efficient technique as it necessitates the acquisition of hundreds of signals for achieving an acceptable signal-to-noise ratio (SNR). To reduce the number of acquisitions and consequently the measurement time, in this study, the BOTDA sensor system is modified to denoise the acquired signals using wavelet shrinkage or adaptive denoising techniques. The application of these denoising techniques to BOTDA sensors shows a significant reduction in the number of averages required to provide an accurate measurement, and consequently a substantial saving of up to 91% in the measurement time of BOTDA sensors.;One of the sources of inaccuracy in BOTDA measurements is the use of curve fitting techniques to estimate the Brillouin frequency shift (BPS) of the points along the fibre. Curve fitting techniques are very sensitive to noise, initialization of fitting parameters, and symmetry, as well as start and stop frequencies. A new method based on the cross-correlation technique is proposed to solve the problem of sensitivity in curve fitting techniques. The results of simulations and experiments show that the proposed method is more robust to noise than curve fitting methods, and it is free from errors associated with the initial settings of fitting parameters.;In addition to the methods introduced to enhance BOTDA measurements, an efficient filter implementation algorithm is designed to optimize the resource requirements for the realization of BOTDA sensors. The filter implementation using the designed algorithm saves up to 22% of the hardware implementation resources compared to conventional algorithms.;The measurement time reduction and the added accuracy achieved in this study enable the implementation of fast and accurate measurements with BODTA sensors and open opportunities to target a new range of dynamic applications.