DSP based implementation of alertness level estimation

To keep on the daily activities, human being need to sleep a certain time everyday. Human spend about one third of his life in sleep and so that sleep is indispensable necessary for life.The aim of this study is estimating the sleep-alertness level from electrical signals taken from brain as DSP based.For this aim, EEG signals taken from 8 healthy subjects were separated as alert, drowsy, and sleep signals in the form of 5 s epochs with the aid of expert doctor. The wavelet coefficients (feature vector) of each EEG signals were obtained by using Discrete Wavelet Transform. Statistical operations were applied to reduce size of feature vectors and obtained vectors were used as input feature vectors of multilayer neural network. The designed Simulink model for classification process was run on TMS320C6713 DSK.The total classification accuracy of proposed model showed that the developed model can be used in the classification of alertness level.