Open AccessExtraction of EEG signals using the discrete wavelet transforms
Author(s) -
H Hindarto,
Arif Muntasa
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/674/1/012055
Subject(s) - electroencephalography , artificial intelligence , pattern recognition (psychology) , computer science , feature extraction , wavelet , cursor (databases) , wavelet transform , brain–computer interface , backpropagation , signal (programming language) , speech recognition , artificial neural network , psychology , psychiatry , programming language
This study focuses on feature extraction for Electro Encephalo Graph (EEG) signals using the Discrete Wavelet Transform method. The EEG signal is used to move up the cursor and the down cursor. In each subband of the EEG signal wave the average value is taken to characterize the EEG signal. Backpropagation Neural Network is used as an EEG signal classification to determine whether the up cursor or the down cursor. The data used in this study are EEG data derived from BCI competition 2003 (BCI Competition 2003). Decision-making is done in two stages. In the first stage, the mean value of each wavelet subband is used as a feature extraction of the EEG signal data. This feature is an input to the Backpropagation Neural Network. In the second stage of the classification process into two classes of class 0 (for the up cursor) and class 1 (for the down cursor), there are 260 training data file of EEG and 293 signals from EEG signal data testing file, so the whole becomes 553 data files of EEG signals. The result obtained for EEG signal classification is 77.2% of the tested signal data.