Decomposition of Sets with Small Boundary (Q4024913)

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 120145
Language Label Description Also known as
default for all languages
No label defined
    English
    Decomposition of Sets with Small Boundary
    scientific article; zbMATH DE number 120145

      Statements

      Decomposition of Sets with Small Boundary (English)
      0 references
      0 references
      16 February 1993
      0 references
      equidecomposable bounded measurable sets
      0 references
      entropy index
      0 references
      discrepancy
      0 references
      Let \(A\) and \(B\) be two subsets of \(\mathbb{R}^k\). \(A\) and \(B\) are said to be equidecomposable using translations, in symbols \(A\overset{tr}\sim B\), if \(A\) can be decomposed into finitely many disjoint subsets that can be translated to give a decomposition of \(B\) into disjoint sets. It is well-known that, for measurable sets \(A\) and \(B\), \(A\overset{tr}\sim B\) implies equal measure, i.e. \(\lambda(A)=\lambda(B)\), \(\lambda\) the Lebesgue measure on \(\mathbb{R}^k\) [see, for example, \textit{S. Wagon}, The Banach-Tarski paradox. Cambridge etc.: Cambridge University Press (1986; Zbl 0569.43001)]. The author shows: if \(A\) and \(B\) are bounded measurable sets with \(\lambda(A)=\lambda(B)>0\), and if the boundaries are small: \(\Delta(\partial A)<k\), \(\Delta(\partial B)<k\), then \(A\overset{tr}\sim B\). Here \(\Delta\) denotes the upper entropy index of \textit{A. N. Kolmogorov} [Dokl. Akad. Nauk SSSR 108, 385--388 (1956; Zbl 0070.11501)]. Corollaries are: (i) If \(A\) and \(B\) are bounded convex sets of equal positive measure, then \(A\overset{tr} \sim B\), and (ii), if \(A\) is a bounded measurable subset of \(\mathbb{R}^k\) of positive measure with \(\Delta(\partial A)<k\), then \(A\) is equidecomposable into a cube of the same measure. In the case of plane sets, even stronger results are possible. Finally, the author answers a question of C. A. Rogers (see the book of S. Wagon, pp. 119 and 230). The author's method is based on a discrepancy argument. The latter allows to formulate a sufficient condition for equidecomposability in terms of the discrepancy of certain sequences.
      0 references

      Identifiers