Classification of Characteristic Waves of Sleep EEG Using First and Higher Order Statistics and Neural Network

Sleep EEG is widely used in the classification of sleep stages, diagnosis of sleep disorders, study of addiction to habits etc.Sleep EEG is one of the most complex signals to analyze. The four characteristic waves, alpha waves, spindle, K-complex and delta waves, of the sleep EEG are important in such applications. In this paper, the classification of these four characteristic waves-alpha waves, spindle, K-complex and delta waves-of sleep EEG using the first and higher order statistics-mean square amplitude, mean slope rate, power spectra and bispectra-and a simple Multi Layer Feed Forward Neural Network, popularly known as Back Propagation Neural Network, has been discussed.