Abstract: Nonlinear dynamic signal processing is attracting several researchers owing to its complex behavior which may be deterministic at macro level and may be in order but unruly behavior with respect to time is difficult to understand and interpret. EEG signals fall under such categories. There are many methods uses for feature extraction e.g. correlation dimension, lyapunov exponent, wavelet transform etc. This paper gives a neuro fuzzy approach to the modeling on EEG signals data in presence of chaos if any. In local modeling approaches, the independent models which work on different nonlinear systems and processes are very successful in modeling, identification, and prediction applications. Chaotic time series are therefore used in our analysis. The results thus produced give a meager prediction error which is desirable to get an efficient analogy to create a much better prediction model for chaotic neuro fuzzy or adaptive neural network systems.

Keywords: EEG signals, feature extraction,Correlation Dimension,Lyapunov Exponent,ANFIS Model.