Pattern recognition for non-invasive EEG-based BCI

Töö põhieesmärk on uurida aina populaarsemaks muutuval EEG tehnoloogial baseeruva Aju-Arvuti liidese võimalusi. Viimaste aastate jooksul on tu- rule jõudnud seadmed, mis on loodud spetsiaalselt tavakasutajate jaoks. Reklaami järgi võimaldavad nad kontrollida arvutit mõttega. Kuid tege- likkuses, selleks, et teha tehnoloogiat töökindlamaks, üritavad need seadmed lugeda ka inimese miimikat, pulssi ja muud sellist. Meid huvitas aga just see, kui palju on võimalik saavutada ainult ajusignaalide lugemisega.We had a look at EEG devices and EEG based BCI technology, which is becoming more and more popular these days. During last few years devices which are designed for wide range of users appeared on the market. Ac- cording to advertisements, these device could allow human to control the computer with power of thought. But in the real applications, to make technology more functional, these devices read not only brainwaves, but also mimics, pulse and other biometrical parameters. We were interested in how much can we achieve if we use only brainwave signal processing and classification