Dimension bounds for invariant measures of bi-Lipschitz iterated function systems

From MaRDI portal



Abstract: We study probabilistic iterated function systems (IFS), consisting of a finite or infinite number of average-contracting bi-Lipschitz maps on R^d. If our strong open set condition is also satisfied, we show that both upper and lower bounds for the Hausdorff and packing dimensions of the invariant measure can be found. Both bounds take on the familiar form of ratio of entropy to the Lyapunov exponent.


This work presents bounds for both the Hausdorff and packing dimensions of invariant measures for probabilistic iterated function systems. Such systems first appeared in the paper [\textit{M. F. Barnsley} and \textit{S. Demko}, Proc. R. Soc. Lond., Ser. A 399, 243--275 (1985; Zbl 0588.28002)], and since then have received much attention from several authors in hundreds of publications.NEWLINENEWLINEThe author presents good results for iterated systems, which include both finite and infinite (but countable) number of maps in \(\mathbb{R}^d\), under the assumptions that the maps are bi-Lipschitz and average-contracting. In particular, the results can be applied to the case where there exists an invariant measure for the iterated system, that can be defined as the fixed point of the Markov operator.NEWLINENEWLINETo prove the results, the concept of \textit{strong open set condition} is introduced, and it gives a way to simplify the geometry of overlapping maps, and it is an essential tool to provide the required bounds.NEWLINENEWLINEIt is a well-written paper, with interesting and useful results, that is worth the read.











This page was built for publication: Dimension bounds for invariant measures of bi-Lipschitz iterated function systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q289547)