Robust γ-filter using support vector machines

Abstract:

This Letter presents a new approach to time-series modelling using the support vector machines (SVM). Although the γ -filter can provide stability in several time-series models, the SVM is proposed here to provide robustness in the estimation of the γ -filter coefficients. Examples in chaotic time-series pbkp_rediction and channel equalization show the advantages of the joint SVM γ -filter. © 2004 Elsevier B.V. All rights reserved.

Año de publicación:

2004

Keywords:

• channel equalization
• γ-Filter
• SUPPORT VECTOR MACHINES
• Iterated pbkp_rediction

Fuente:

scopus