Possibilistic entropy: A new method for nonlinear dynamical analysis of biosignals

Abstract

The theory of nonlinear dynamical systems has opened doors to discovering potential patterns hidden in complex time-series data. An attrative approach to nonlinear time-series analysis is the measure of predictability which characterizes the data in terms of entropy. A new entropy measure is presented in this paper as a new nonlinear dynamical method, which is based on the theory of possibility and the kriging computation. The proposed model has the potential for studying complex biosignals. © 2011 Springer-Verlag.