1
EEG Signal Classification for Brain
Computer Interface Applications
ECOLE POLYTECHNIQUE FEDERALE DE LAUSANNE
Jorge Baztarrica Ochoa
Responsible Assistant : Gary Garcia Molina.
Professor : Touradj Ebrahimi
March 28th, 2002
2
Abstract
Recent advances in computer hardware and signal processing have made possible
the use of EEG signals or “brain waves” for communication between humans and
computers. Locked-in patients have now a way to communicate with the outside world,
but even with the last modern techniques, such systems still suffer communication rates
on the order of 2-3 tasks/minute. In addition, existing systems are not likely to be
designed with flexibility in mind, leading to slow systems that are difficult to improve.
This diploma project explores the effectiveness of Time – Frequency Analysis as
a
technique of classifying different mental
tasks
through the use of
the
electroencephalogram (EEG). EEG signals from several subjects through 6 channels
(electrodes) have been studied during the performance of five mental tasks (a baseline
resting task, mental multiplication, geometric figure rotation, mental letter composition,
and counting). Improved off-line classification of two of them (“geometric figure
rotation” and “mental letter composition”), for which poor results had been obtained with
autoregressive models before, were the principal objective of this project.
Different methods based on Time Frequency Representations have been
considered for the classification between the two tasks mentioned above. A non-iterative
method based on the Ambiguity Function was finally selected. The results indicate that
this method is able to extract in half-second, distinguishing features from the data, that
could be classified as belonging to one of the two tasks with an average percentage
accuracy which tends to zero. The same results were found when the method was
exported for five tasks EEG signal class