scientific article; zbMATH DE number 1179913
An iterated function system (IFS) on a metric space \((X,d)\) is a family \((S_i)_{i\in I}\) of continuous transformations of \(X\). Denoting by \(\mathcal F\) the class of all nonempty closed subsets of \(X\), the Barnsley operator associated to an IFS is a mapping \(F:{\mathcal F}\to {\mathcal F}\) defined by: \[ F(A)=\overline{\bigcup_{i\in I}S_i(A)}, \forall A\in {\mathcal F} \] An IFS is asymptotically stable if there exists a set \(A_0\in{\mathcal F}\) such that for every bounded set \(A\in {\mathcal F}\), \(A_0\) is the topological limit of the sequence \((F^n(A))\). The set \(A_0\) is called the attractor of the IFS. It is an \(F\)-invariant set. If a subsystem \((S_i)_{i\in I_0}\) is asymptotically stable then its attractor \(A_0\) is called a nucleus of the original system. The set \(A*=\bigcup_{n=1}^\infty F^n(A_0)\) associated to the nucleus \(A_0\) is called the semifractal corresponding to the IFS. The aim of the paper is to give a method of construction of semifractals on Polish spaces. If \((X,d)\) is a Polish space denote by \({\mathcal M}\), the family of all finite measures on \(X\) and by \({\mathcal M}_s\), the space of all finite signed measures. One defines an IFS with probabilities as a family of pairs \((S_i,p_i)_{i\in I}\), where \(p_i:X\to [0,1]\) are continuous functions such that \(\sum_{i\in I}p_i(x)=1\), \(\forall x\in X\). To an IFS with probabilities one associates the Markov operator \(P:{\mathcal M}_s\to {\mathcal M}_s\) defined by: \[ P\mu(A)=\sum_{i\in I}\int_{S_i^{-1}(A)}p_i(x)\mu dx, \forall \mu\in {\mathcal M}_s, \text{and} \forall A, \text{ Borel subset of } X \] The asymptotic stability for Markov operators is defined. In this context an IFS with probabilities is called asymptotically stable if the corresponding Markov operator has this property. Sufficient conditions for asymptotic stability of the Markov operator are derived. Moreover, conditions on \(S_i\) that ensure the existence of probabilities \(p_i\) such that the IFS \((S_i,p_i\)) is asymptotically stable are also found. Finally, it is given an effective way for the construction of semifractals using Markov chains.
- Multivalued fractals
- On existence and structure of semiattractors for dynamical systems represented by cocycles
- Some New Concepts of Dimension
- Random iteration for infinite nonexpansive iterated function systems
- scientific article; zbMATH DE number 887295 (Why is no real title available?)
- FRACTALS, SEMIFRACTALS AND MARKOV OPERATORS
- Minimal invariant closed sets of set-valued semiflows
- On the geometric ergodicity for a generalized IFS with probabilities
- Semifractals from multifunctions on product spaces and generalized iterated function systems
- Strongly-Fibred Iterated Function Systems and the Barnsley–Vince triangle
- Fractal approximation for phreatic semivariograms
- Asymptotic properties of multifunctions, families of measures and Markov operators associated with cocycles
- Semiattractors of set-valued semiflows
- How to construct asymptotically stable iterated function systems
- An ergodic decomposition defined by transition probabilities
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4399192)