Parameter estimation for filtered discrete fractal signals

Abstract

The problems of estimating the parameters of discrete fractals and filtered discrete fractals directions are considered. A maximum likelihood approach for estimating the parameters of pure fractals is presented. In particular, it is shown that the computation of the likelihood function does not require any matrix inversion as previously thought. For filtered fractals, an iterative procedure is proposed to estimate the parameters of the fractal signal and those of the shaping filter. The procedure has been shown to converge experimentally.<>