Biomedical signal processing with time-varying CML model

Brain electrical signals, one of the typical biomedical signals, is accepted to be produced by the nonlinear dynamic system. In brain electrical signal, both EEG and ERP signal is consider as chaotic signal. In this paper, a new time variation based coupled map lattice(CML) model is put forward to research the nonlinear character of the ERP signal under specified tasks. To further investigate the ERP model and discover new information, the time variation largest Lyapunov exponent (LLE) is developed. Both simulated EEG and acquired ERP experiments were used to test. And the results show that the brain chaos changes corresponding to the changes of attention task. There is an optimum scopes size of cues in the visual attention information processing.