Fractional fuzzy entropy algorithm and the complexity analysis for nonlinear time series

In this paper, fractional fuzzy entropy (FFuzzyEn) algorithm is designed by combing the concept of fractional information and fuzzy entropy (FuzzyEn) algorithm. Complexity of chaotic systems is analyzed and parameter choice of FFuzzyEn is investigated. It also shows that FFuzzyEn is effective for measuring dynamics of nonlinear time series and has better comparing results for different time series. Moreover, changes in the complexity of EEG signals from normal health persons and epileptic patients are observed. The results show that, compared with normal health persons, epileptic patients have the lowest complexity during seizure activity and relative lower complexity during seizure free intervals. The proposed method may be useful for EEG signal based physiological and biomedical analysis.