Abstract
A curve-fitting procedure based on the simulated annealing algorithm has been developed for the analysis of spectral Raman data. By the inclusion of a priori information about the instrumental broadening in the definition of the cost function that is minimized, effects of the finite instrumental resolution are eliminated from the resulting fit. The ability of the method to reproduce original band shapes is tested on synthesized spectra and FT-Raman spectra of diamond recorded at different resolutions with different apodization functions. The procedure yields the global optimum of the fitted parameters and is easily implemented on a personal computer.