Benko Tina P, Perc Matjaz
Department of Physics, Faculty of Natural Sciences and Mathematics, University of Maribor, SI-2000 Maribor, Slovenia.
Biosystems. 2009 Sep;97(3):154-9. doi: 10.1016/j.biosystems.2009.05.011. Epub 2009 Jun 6.
We use nonlinear time series analysis methods to analyze the dynamics of the sound-producing apparatus of the American crocodile (Crocodylus acutus). We capture its dynamics by analyzing a recording of the singing activity during mating time. First, we reconstruct the phase space from the sound recording and thereby reveal that the attractor needs no less than five degrees of freedom to fully evolve in the embedding space, which suggests that a rather complex nonlinear dynamics underlies its existence. Prior to investigating the dynamics more precisely, we test whether the reconstructed attractor satisfies the notions of determinism and stationarity, as a lack of either of these properties would preclude a meaningful further analysis. After positively establishing determinism and stationarity, we proceed by showing that the maximal Lyapunov exponent of the recording is positive, which is a strong indicator for the chaotic behavior of the system, confirming that dynamical nonlinearities are an integral part of the examined sound-producing apparatus. At the end, we discuss that methods of nonlinear time series analysis could yield instructive insights and foster the understanding of vocal communication among certain reptile species.
我们使用非线性时间序列分析方法来分析美国短吻鳄（Crocodylus acutus）发声器官的动力学。我们通过分析交配期歌唱活动的录音来捕捉其动力学。首先，我们从声音记录中重建相空间，从而揭示吸引子在嵌入空间中充分演化需要不少于五个自由度，这表明其存在背后存在相当复杂的非线性动力学。在更精确地研究动力学之前，我们测试重建的吸引子是否满足确定性和平稳性的概念，因为缺乏这些属性中的任何一个都会妨碍有意义的进一步分析。在确定了确定性和平稳性之后，我们接着表明录音的最大李雅普诺夫指数为正，这是系统混沌行为的一个有力指标，证实了动态非线性是所研究发声器官的一个组成部分。最后，我们讨论非线性时间序列分析方法可以产生有启发性的见解，并促进对某些爬行动物物种之间声音交流的理解。
