Common spatial patterns based on generalized norms

Author

Jangwoo Park ; Wonzoo Chung

Author_Institution

Dept. of Comput. & Commun. Eng., Korea Univ., Seoul, South Korea

fYear

2013

fDate

18-20 Feb. 2013

Firstpage

39

Lastpage

42

Abstract

The Common Spatial Patterns (CSP) algorithm is commonly used to finds spatial filters for classification of electroencephalogram (EEG) signals. However, conventional CSP is sensitive to outliers and artifacts because it is based on variance using L2-norm. In this paper, we consider generalized L_p norm based CSP, called CSP-L_p, and verify whichp is optimal for CSP-L_p by maximizing the Lp norm ratio of filtered dispersion of one class to the other class. The spatial filters of CSP-L_p are obtained empirically. Simulation result on a toy example shows the robustness of CSP-L_p depending on L_p-norm.

Keywords

brain-computer interfaces; electroencephalography; filtering theory; medical signal processing; signal classification; statistical analysis; CSP algorithm; EEG signal classification; L2-norm; common spatial patterns algorithm; electroencephalography; generalized norm; spatial filter; variance; Classification algorithms; Covariance matrices; Dispersion; Electroencephalography; Robustness; Signal processing algorithms; Spatial filters; CSP-Lp; Common spatial patterns (CSP); Lp-norm; brain computer interfaces (BCI); electroencephalogram (EEG) classification;

fLanguage

English

Publisher

ieee

Conference_Titel

Brain-Computer Interface (BCI), 2013 International Winter Workshop on

Conference_Location

Gangwo

Print_ISBN

978-1-4673-5973-3

Type

conf

DOI

10.1109/IWW-BCI.2013.6506623

Filename

6506623