Analysis of Asynchronous Brain Computer Interface Using Discrete Wavelet Transform


AKMAN AYDIN E. , BAY Ö. F. , GÜLER İ.

26th IEEE Signal Processing and Communications Applications Conference (SIU), İzmir, Türkiye, 2 - 05 Mayıs 2018 identifier identifier

Özet

Brain Computer Interfaces (BCIs) which offer a brain-computer-actuator communication channel as an alternative to brain-nerve-muscle-activator communication channel have been studied intensively in recent years. Asynchronous BCIs (A-BCI) determine whether users interact with the BCI system, and produce an output command only in the case determining the subjects' control state. In this study, performance of the A-BCI system is analyzed by using Discrete Wavelet Transform (DWT) method; performance of the system is evaluated in terms of control state detection accuracy and the number of intensification sequences required for selection. According to the results, the proposed method determines the control and non-control states with an accuracy of 99.66%. Mean number of intensification sequences to make selections at control state is determined as 3.55.