Next: Experimental Setup
Up: A Sliding-Window Kernel RLS
Previous: Updating the Inverse of
Example Problem: Identification of a Nonlinear Wiener System
In this section, we consider the
identification problem for a nonlinear Wiener system, and compare
the performance of the proposed kernel RLS algorithm to the
standard approach using a multilayer perceptron (MLP). Since
kernel methods provide a natural nonlinear extension of linear
regression methods, the proposed system is supposed to perform
well compared to the MLP [7].
Subsections
Pdf version (187 KB)
Steven Van Vaerenbergh
Last modified: 2006-03-08