I'm working with self-similar traffic in computer networks. There are
some traffic traces that are multifractals, i.e. the hurst exponent
varies in time. Many papers uses the wavelets to estimate the Holder
exponent. I would like to know if there is other methods to do this.
Thanks,