ADAPTIVE RECURRENT NEURAL NETWORK FOR REDUCTION OF NOISE AND ESTIMATION OF SOURCE FROM RECORDED EEG SIGNALS
ARPN Journal of Engineering and Applied Sciences, Vo. 10, No.3, Februari 2015, ISSN 1819-6608
Selasa, 23 Mei 2017 11:25 | Sudah dibaca 136 kali
In recording the EEG signals are often contaminated by a large of signals called artifacts such that the brain activity (source) difficult to estimate. There are different kinds of artifacts such as power line noise, electromyogram, electrocardiogram and electrooculogram. In this research, an adaptive recurrent neural network (ARNN) for estimation of source and reduction of noise from recorded EEG signals is proposed. In the experiment, the EEG signals are recorded on three conditions, which is normal conditions, closed eyes, and blinked eyes. After processing, the dominant frequency of the EEG signal is obtained in the range of 12-14 Hz either on normal conditions, closed eyes, and blinked eyes. The experimental results show that the ARNN method was effectively estimated the brain activity according to the given stimulus and remove the artifacts from all subjects.
Kata Kunci: EEG, adaptive, neural network, estimation, reduction.