Brainwave-controlled applications with the Emotiv EPOC using support vector machine

Electroencephalography (EEG) which is the electrical signal recorded by the sensors attached on the human scalp to detect brain activities has been the emerging trends in digital signal processing. As compared to processing other types of digital data such as speech or audio signals, EEG signal processing is more challenging. However, EEG signals have practically found a wide range of important applications. In this paper, we propose a design of a brain-computer interface (BCI) using EEG's P300 component to a control application. First, we use the Emotiv EPOC headset to capture the raw EEG signals. Then, we adopt a classification algorithm by invoking support vector machine along with the selected extracted features to classify the two-class EEG trials (with and without P300 component). The algorithm is developed to help people express their selection of one among four commands. The experimental results are provided evaluate the classification accuracy.