Total variation denoising and Euclidean distance based distributed temperature monitoring in Brillouin optical time-domain analysis sensors

Abstract

The performance of Brillouin optical time domain analysis (BOTDA) sensors is largely deteriorated due to the poor signal-to-noise ratio (SNR) of Brillouin gain spectra (BGSs) collected from the BOTDA experiment. The fast monitoring of distributed temperature using BOTDA sensors is also vital for many longdistant applications. To cope with these requirements, this paper proposes total variation denoising (TVD) and Euclidean distance-based pattern recognition (TEPR) for high-performance BOTDA sensors. The performances of TEPR are analyzed explicitly, and rigorous comparisons have been made with traditional nonlinear least squares fitting (NLSF). The experimentally demonstrated results signify that the proposed TEPR can improve the measurement uncertainty by up to ~55% compared to NLSF without worsening the experimental spatial resolution. The signal processing for using TEPR is also ~4 times faster than that for using NLSF. Hence, the proposed technique is an efficient and reliable alternative for the fast and accurate monitoring of distributed temperature in BOTDA sensors. J.

Bangladesh Acad. Sci. 46(2); 193-202: December 2022