In this study, electroencephalogram (EEG) signals recorded during transition from wakefulness to sleep and Matlab-Simulink and TMS320C6713 DSP Starter Kit (DSK) of Texas Instruments Inc. are used for classification of alertness level. First, 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 subbands(feature vector) of each EEG signals were obtained by using Discrete Wavelet Transform. Some statistical operations were used to reduce dimensions of feature vectors and obtained vectors were chosen as input feature vectors of multilayer neural network which is used as classifier. The Simulink model for real time classification process was run on DSK. The tests showed that the results of classification with DSK are same with the results of classification simulation without using DSK. Total classification accuracy obtained in the test results of proposed model showed that the model can be used in the estimation of alertness level.
[1]
Elif Derya Übeyli,et al.
Neural network analysis of internal carotid arterial Doppler signals: predictions of stenosis and occlusion
,
2003,
Expert Syst. Appl..
[2]
E. Wolpert.
A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects.
,
1969
.
[3]
H. Adeli,et al.
Analysis of EEG records in an epileptic patient using wavelet transform
,
2003,
Journal of Neuroscience Methods.
[4]
Necmettin Sezgin,et al.
The ANN-based computing of drowsy level
,
2009,
Expert Syst. Appl..
[5]
Ah Chung Tsoi,et al.
Classification of EEG signals using the wavelet transform
,
1997,
Proceedings of 13th International Conference on Digital Signal Processing.
[6]
Rulph Chassaing.
DSP Applications Using C and the TMS320C6x DSK
,
2002
.