Berita BPI LIPI
oleh : , M. Agung Suhendra; ; Poltak Sihombing
In this paper, an extraction and classification of steady state-visual evoked potentials using the IIR Chebyshev I of 4 order and the adaptive feed-forward Neural Networks algorithm, respectively are applied. The classification results of the extracted signals is directly used to make a user able of controlling the directions (stop, forward, right, and left with stimuli frequencies of 7.5, 10, 15, and 20 Hz, respectively) of a wheelchair based brain computer interface. The data was collected during a session in which fourteen subjects with age about 24±2 years were tested. The average classification accuracy level around 82% of four directions is achieved. It is improve about 13.75% compare with the obtained previous result.