Discrimination of alcohol and normal EEG signal using EMD

Abstract

The discrimination of alcoholic and normal person using electroencephalogram (EEG) signal is a challenging task in the field of biomedical research. To solve this task, this research work uses Empirical mode decomposition (EMD) method for the discrimination of alcoholic and normal person using EEG signal. The EMD is very useful method to deal with non-linear and non-stationary EEG signals, it decompose the EEG signal into set of narrow-band intrinsic mode functions (IMFs). The obtained IMFs with Hilbert transform is used to compute following features: amplitude modulation bandwidth (B am ), frequency modulation bandwidth (B Fm ), mean of instantaneous frequency (fi mean ) and Entropy. The discrimination ability of these features is obtained by Kruskal-Wallis (KW) statistical test. The lower probability value shows that the features have statistically significant for discrimination of alcohol and normal EEG signals.