Adaptive filtering using projection onto convex sets

LEONARDO REY VEGA; SARA TRESSENS; HERNÁN REY

Río Cuarto, Argentina

Conferencia; Reunión en Procesamiento de la Información y Control RPIC 2005; 2005

Universidad Nacional de Río Cuarto

In this paper we propose a novel adaptive filtering algorithm. The algorithm exploits the information given by the power spectral density of the noise extracted from  the periodogram of filtering error. The goal is try to match the spectral properties of the error  filtering with the spectral properties of the measurement noise. With this in mind appropriate convex  and closed sets are built and projections onto them are computed. The simulations results show that   the algorithm has excellent convergence properties with a reduced number of updates. This could be exploited to obtain a lower computational load.