Novel data analysis approach for temperature and strain profile reconstruction in distributed fiber optics sensors based on Stimulated Brillouin Scattering

The reconstruction of temperature or strain profiles along an optical fiber in Brillouin scattering-based systems, is usually made by directly reading the Brillouin power spectra extracted for each fiber location. Such approach suffers from systematic errors due mainly to nonlocal effects. In this paper, we propose a novel numerical technique for temperature/strain profile reconstructions based on Brillouin optical-fiber time-domain analysis (BOTDA) sensors. In this approach we search directly for the temperature/strain profile along the fiber that matches the measured data. The algorithm is based on an harmonic expansion of the unknown profile, whose coefficients are determined by means of a multidimensional minimization. Several numerical simulations, even in presence of noise, have proved the capability of the proposed algorithm to compensate for systematic errors suffered by classical approaches.