SciELO - Scientific Electronic Library Online

 
vol.22 número6Sisal: a supervisory system for oil wellsCompact yet efficient hardware architecture for multilayer-perceptron neural networks índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Sba: Controle & Automação Sociedade Brasileira de Automatica

versión impresa ISSN 0103-1759

Resumen

BENEVIDES, Alessandro B.; SARCINELLI-FILHO, Mário  y  BASTOS FILHO, Teodiano F.. Design of a general brain-computer interface. Sba Controle & Automação [online]. 2011, vol.22, n.6, pp. 638-646. ISSN 0103-1759.  http://dx.doi.org/10.1590/S0103-17592011000600009.

This paper presents the classification of three mental tasks, using the EEG signal and simulating a real-time process, what is known as pseudo-online technique. The Bayesian classifier is used to recognize the mental tasks, the feature extraction uses the Power Spectral Density, and the Sammon map is used to visualize the class separation. The choice of the EEG channel and sampling frequency is based on the Kullback-Leibler symmetric divergence and a reclassification model is proposed to stabilize the classifications.

Palabras llave : Brain-Computer Interface; Power Spectral Density; Kullback-Leibler symmetric divergence.

        · resumen en Portugués     · texto en Inglés     · pdf en Inglés