l1 - NORM REGULARIZATION FOR SPARSE REPRESENTATION AND P300 WAVE DETECTION IN B RAIN -COMPUTER INTERFACES

VICTORIA PETERSON; HUGO LEONARDO RUFINER; RUBÉN SPIES

Tandil

Congreso; V MACI 2015; 2015

ASAMACI

Brain-Computer Interface (BCI) is a system which provides direct communication between the mindof a person and the outside world by using only brain activity (EEG). A common EEG-BCI paradigm is based onthe so called Event-Related Potentials (ERP) which are responses of the brain to some external stimuli. One ofthe main components of ERP signals is an enhanced positive-going component called P300 wave. The ` 1 -normminimization has been widely used due to its sparsity-inducing property, convenient convexity and great success inseveral applications. In this work we propose a sparse representation and posterior classification of ERPs signals bymeans of an ad-hoc spatio-temporal dictionary composed of bidimensional Gaussian elements. The classification isbased on minimizing the residual between a test sample and its estimation.