Analysis of dynamic light scattering signals with discrete wavelet transformation

The analysis of non-stationary signals calls for specific tools which go beyond classical Fourier analysis. Here we present the discrete wavelet transform as a quantitative method to analyse the scaling properties of non-stationary and chaotic signals obtained by light scattering. The analysis is performed directly on the time-resolved scattered intensity. The result is the scaling exponent of the underlying power spectrum and the frequency window where the scaling can be observed. These properties are obtained with a higher accuracy and in a fraction of the time compared with the classical method of intensity correlation functions. As a test we applied the new method to the sol-gel transition of a suspension of disc-shaped colloidal particles in water.