Characterization of chaotic dynamics in the vocalization of Cervus elaphus corsicanus

Abstract

Chaos, oscillations, instabilities, intermittency represent only some nonlinear examples apparent in natural world. These phenomena appear in any field of study, and advances in complex and nonlinear dynamic techniques bring about opportunities to better understand animal signals. In this work we suggest an analysis method based on the characterization of the vocal fold dynamics by means of the nonlinear time series analysis, and by the computations of the parameters typical of chaotic oscillations: Attractor reconstruction, Spectrum of Lyapunov Exponents and Maximum Lyapunov Exponent was used to reconstruct the dynamic of the vocal folds. Identifying a sort of of vocal fingerprint can be useful in biodiversity monitoring and understanding the health status of a given animal. This method was applied to the vocalization of the Cervus elaphus corsicanus, the Sardinian Red Deer.