A novel fitting algorithm for Brillouin scattering spectrum of distributed sensing systems based on RBFN networks

In order to extract features for Brillouin scattering spectrum of distributed sensing systems with high accuracy, a novel fitting algorithm, using a hybrid algorithm based on Particle Swarm Optimization algorithm and Levenberg-Marquardt algorithm to optimize the optimization process of radial basis function networks, is explanatorily proposed. Compared the proposed algorithm with traditional BP neural networks, the five times polynomial curve and piecewise cubic spline interpolation in fitting the simulative and experimental spectrum, respectively, the evaluation parameter is relatively better than other three algorithms under the same experiment with different pulse widths. The numerical and experimental results showed that modified RBFN networks have some referential roles, which can guarantee the accurate measurement of Brillouin scattering spectrum.