A NEW METHOD TO ESTIMATE THE KOLMOGOROV-ENTROPY FROM RECURRENCE PLOTS- ITS APPLICATION TO NEURONAL SIGNALS

Recurrence plots are a useful graphic tool for studying dynamical syst ems. A new variable is proposed to quantify their deterministic struct ure. It allows an easy computation of the KZ entropy since the distanc e between points needs to be calculated only once instead of every tim e for each new embedding dimension, and it bypasses some of the Limita tions inherent to the classical Grassberger and Procaccia method. It i s valuable for biological data, as illustrated here by tests with vari ous dynamical models and its application to experimental signals recor ded from neurons. (C) 1998 Elsevier Science B.V.