Abstract
The development of neurocomputer interfaces (NCI) using the EEG requires use of effective algorithms for signal analysis. One approach to creating NCI is based on use of the characteristics of individual visual event-related potentials (VEP) for control purposes. However, this is a difficult task requiring a combination of different approaches related to signal processing, particularly blind decomposition of sources (blind source separation), machine learning, and various others. We present here results from a comparative analysis of several classifiers in a task consisting of recognition of individual VEP. Open-access EEG traces were used for this study.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
