Raman spectroscopy can be used in situ real-time measurement because it's rapid, and it is helpful to real-time online monitoring of process control. With the complexity of the environment and the characteristics of Raman signal, it is hard to avoid some overlapping spectrum peaks. Based on the advantage of immune algorithm, an immune algorithm (IA) was applied to the overlapping Raman signals of aromatics. With extraction of each single Raman spectrum peak signal from the mixture signals for resolution, Results show that the method is effective to identify the overlapped Raman signal for its fast resolution and accurate quantitative determination with the relative error less than 1%. For the overlapping Raman signals with fluorescence background disturbance, we proposed an adaptive immune algorithm, which is combined with independent component analysis. It can effectively resolve the fluorescence background signal, and it provides a new way for Raman spectra analysis of complex samples.