Bispectral analysis, a technique based on high-order statistics, is extended to encompass time dependence for the case of coupled nonlinear oscillators. It is applicable to univariate as well as to multivariate data obtained, respectively, from one or more of the oscillators. It is demonstrated for a generic model of interacting systems whose basic units are the Poincaré oscillators. Their frequency and phase relationships are explored for different coupling strengths, both with and without Gaussian noise. The distinctions between additive linear or quadratic, and parametric (frequency modulated), interactions in the presence of noise are illustrated.

Abstract. Bispectral analysis based on high order statistics, introduced recently as a technique for revealing time-phase relationships among interacting noisy oscillators, has been used to study the nature of the coupling between cardiac and respiratory activity. Univariate blood flow signals recorded simultaneously by laser-Doppler flowmetry on both legs and arms were analysed. Coupling between cardiac and respiratory activity was also checked by use of bivariate data and computation of the cross-bispectrum between the ECG and respiratory signals. Measurements were made on six healthy males aged 25-27 years. Recordings were taken during spontaneous breathing (20 min), and during paced respiration at frequencies both lower and higher than that of spontaneous respiration (either two or three recordings with a constant frequency in the interval between 0.09 and 0.35 Hz). At each paced frequency recordings were taken for 12 min. It was confirmed that the dynamics of blood flow can usefully be considered in terms of coupled oscillators, and demonstrated that interactions between the cardiac and respiratory processes are weak and time-varying, and that they can be nonlinear. Nonlinear coupling was revealed to exist during both spontaneous and paced respiration. When present, it was detected in all four blood flow signals and in the cross-bispectrum between the ECG and respiratory signal. The episodes with nonlinear coupling were detected in 11 out of 22 recordings and lasted between 19 s in the case of high frequency (0.34 Hz) and 106 s in the case of low frequency paced respiration (0.11 Hz).

In the natural world, the properties of interacting oscillatory systems are not constant, but evolve or fluctuating continuously in time. Thus, the basic frequencies of the interacting oscillators are time varying, which makes the system analysis complex. For studying their interactions we propose a complementary approach combining wavelet bispectral analysis and information theory. We show how these methods uncover the interacting properties and reveal the nature, strength, and direction of coupling. Wavelet bispectral analysis is generalized as a technique for detecting instantaneous phase-time dependence for the case of two or more coupled nonlinear oscillators whereas the information theory approach can uncover the directionality of coupling and extract driver-response relationships in complex systems. We generate bivariate time-series numerically to mimic typical situations that occur in real measured data, apply both methods to the same time-series and discuss the results. The approach is applicable quite generally to any system of coupled nonlinear oscillators.