Detect determinism of spontaneous EEG with a multi-channel reconstruction method

In this paper, a multi-channel reconstruction with correlation coefficients method is present for time series' determinism analysis. Nonlinear measures of determinism and D2 estimator for both single and multi-channel time series are calculated. It shows that the reconstruction method has well performance with noise and an obviously little variety to the selection of time delay and embedding dimension. Using this method, we demonstrate the multi-channels spontaneous EEG signals of five mental tasks exhibit a finite determinism structure which could be used for further nonlinear study. The method is applicable to online EEG analysis.