The origin of diffusion: the case of non-chaotic systems

Abstract

We investigate the origin of diffusion in non-chaotic systems. As an example, we consider 1-d map models whose slope is everywhere 1 (therefore the Lyapunov exponent is zero) but with random quenched discontinuities and quasi-periodic forcing. The models are constructed as non-chaotic approximations of chaotic maps showing deterministic diffusion, and represent one-dimensional versions of a Lorentz gas with polygonal obstacles (e.g., the Ehrenfest wind tree model). In particular, a simple construction shows that these maps define non-chaotic billiards in space-time. The models exhibit, in a wide range of the parameters, the same diffusive behavior of the corresponding chaotic versions. We present evidence of two sufficient ingredients for diffusive behavior in one-dimensional, non-chaotic systems: i) a finite-size, algebraic instability mechanism, and ii) a mechanism that suppresses periodic orbits.

Effect of Meditation on Scaling Behavior and Complexity of Human Heart Rate Variability

“…It is this threat and this promise that is making the science of complex systems as of the fastest growing areas of science at the present time. Recently there have been some attempts to quantify the complexity of the complex systems [21][22][23]. Though there is no formal definition of complexity, it is considered to be a measure of the inherent difficulty to achieve the desired understanding.…”

“…Increase in entropy corresponds to the increase in the degree of disorder and for a completely random system it is maximum. Traditional algorithms are single-scale based [21][22][23][24]. Zhang's method was based on the Shannon entropy which requires a large number of almost noise free data.…”

Brownian motion and diffusion: From stochastic processes to chaos and beyond

“…This is key-point, because someone can argue that a deterministic infinite system with spatial randomness can be interpreted as an effective stochastic system, but this is probably a "matter of taste". With the aim of clarifying this point, we consider now a spatially disordered non-chaotic model 54 , which is the one-dimensional analog of a two-dimensional non-chaotic Lorentz system with polygonal obstacles. It has the advantage that both the case with finite and zero spatial entropy density can be investigated.…”

Macroscopic evidence of microscopic dynamics in the Fermi-Pasta-Ulam oscillator chain from nonlinear time-series analysis

“…Later on it was found that the chaotic Lorentz model, which has circular scatterers, has a diffusive behavior that is indistinguishable from the one exhibited by the Eherenfest model [3]. Furthermore, a class of one-dimensional maps has also been reported which present normal diffusivelike behavior in the absence of chaos, just as in the Eherenfest model [4]. These results have rendered doubts about the conclusion * Electronic address: rbm@xanum.uam.mx that microscopic chaos has been experimentally detected.…”

“…Furthermore, this is a many-particle system in which the statistical behavior of the heavy impurity is produced by the interaction with the oscillators of the chain, in sharp contrast with the models of Refs. [4,5], which are one-dimensional maps with only one dynamical variable, i.e. the position, defined in the model [6].…”