Set theoretic adaptive filltering: using periodogram and projection onto convex sets

LEONARDO REY VEGA; SARA TRESSENS; HERNÁN REY

Antalya, Turquía

Conferencia; European Signal Processing Conference, EUSIPCO; 2005

European Association for Signal Processing, EURASIP

In this paper we propose a novel adaptive filtering algorithm. Using the Set Theoretic Estimation  framework, the algorithm exploits the information given by the power spectral density of the noise extracted from  the periodogram of filtering error. With this information new appropriate sets are built and projections onto them are  computed. The simulations results show that the algorithm has excellent convergence properties.