PaperThe following article is Open access

Topological synchronisation or a simple attractor?

, , and

Published 30 May 2023 © 2023 IOP Publishing Ltd & London Mathematical Society
, , Citation Théophile Caby et al 2023 Nonlinearity 36 3603DOI 10.1088/1361-6544/acd42f

0951-7715/36/7/3603

Abstract

A few recent papers introduced the concept of topological synchronisation. We refer in particular to (Lahav et al 2022 Sci. Rep.12 2508), where the theory was illustrated by means of a skew product system, coupling two logistic maps. In this case, we show that the topological synchronisation could be easily explained as the birth of an attractor for increasing values of the coupling strength and the mutual convergence of two marginal empirical measures. Numerical computations based on a careful analysis of the Lyapunov exponents suggest that the attractor supports an absolutely continuous physical measure (acpm). We finally show that for some unimodal maps such acpm exhibit a multifractal structure.

Export citation and abstractBibTeXRIS

Original content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

Footnotes

  • It is worth noticing that in the next section we will compute the Dq spectrum in a few cases by using the characterization (8) and not the definition (7) in terms of the correlation integral.

  • We point out, however, that it is in general not enough to have a weak convergence of the measures to ensure the convergence of the Dq spectrum. Suppose for instance that the master system has an absolutely continuous invariant measure and that, for k close enough to 1, so does the measure of the slave system . If with as, for instance, it is the case for some quadratic map along the orbit of the critical point, then the local dimension of µ at x0 is and it is easily seen (see the detailed computations in the proof of Proposition 5.1) that the Dq spectrum is not constant. Moreover, if we further assume that, for all k < 1, hk is a piecewise constant function converging in L1 to h, it is easy to see that the Dq spectrum for µk is constant equal to 1 for all k < 1, so that there is no convergence to the spectrum of the master map.

  • The map f is of class C4 and it must be:

    • a Collet-Eckmann S-unimodal map verifying with where H0 is a constant larger than 1.
    • a Benedicks-Carleson map: such that .

  • With the symbol we mean that a is bounded from below and above as with two positive constants.

  • The continuity of T1 and of T2 and therefore of ψ is sufficient.

  • 10 

    In other words we are supposing that the are mutually independent. This is of course not true when ωn is distributed as , with x chosen Leb-a.e. but it becomes asymptotically true since T mixes exponentially fast with respect to µ.

undefined