IFS attractors and Cantor sets (Q2493894)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 5033415
Language Label Description Also known as
default for all languages
No label defined
    English
    IFS attractors and Cantor sets
    scientific article; zbMATH DE number 5033415

      Statements

      IFS attractors and Cantor sets (English)
      0 references
      0 references
      0 references
      16 June 2006
      0 references
      An iterated function system (IFS) is a finite family of contractive maps \(\{f_{1},f_{2},\dots ,f_{n}\}\) acting on a complete metric space. It is well known that such a dynamical system has a unique attractor, that is, a unique compact set \(K\subset X\) which is nonempty and invariant by the IFS, in the following sense: \(K=\bigcup_{i=1} ^{n}f_{i}(K)\). This paper addresses the following question: is every Cantor set an attractor of some iterated function system? For instance, the usual middle one third Cantor set is an attractor of the IFS \(\{f_{1},f_{2}\}\) on the real line defined by \(f_{1}(x)=x/3\) and \(f_{2}(x)=x+2/3\). The authors show that for some other metrics, the Cantor set is no more the attractor of an IFS. Theorem: There exists a Cantor set \(X_{1}\) and a Borel probability measure \(\mu\) supported on \(X_{1}\) such that for every contractive map \(f\) on \(X_{1}\), we have \(\mu (f(X_{1}))=0\). Moreover, for any iterated function system on a complete metric space \((X,d)\), the attractor \(K\) is not isometric to the Cantor set \(X_{1}\). The authors prove that is also possible to find topological impediments for \(K\) to be an attractor of an IFS. Theorem: There exists a Cantor set \(X_{2}\subset {\mathbb R}^{3}\) such that if \(f\) is a homeomorphism of \({\mathbb R}^{3}\) which satisfies \(f(X_{2})\subset X_{2}\), then \(f| _{X_{2}}\) is the identity. In particular, a finite set of homeomorphisms of \({\mathbb R}^{3}\) can not be an IFS whose attractor is \(X_{2}\). (The set \(X_{2}\) here is a variation on Antoine's necklace.) Although the foregoing result can be generalized to higher dimensions, it is interesting to point out the following result presented in this paper. Theorem: For any Cantor set \(X\) in \({\mathbb R}\) (or in \({\mathbb R}^{2}\)) and any two points \(x,y\in X\), there exists a contractive homeomorphism \(f\) on \({\mathbb R}\) (or on \({\mathbb R}^{2}\)) such that \(f(X)\subset X\) and \(f(x)=y\).
      0 references
      Cantor set
      0 references
      iterated function system
      0 references
      Antoine's necklace
      0 references
      0 references
      0 references
      0 references

      Identifiers