No label defined (Q4812295)

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 2097540
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 2097540

      Statements

      7 September 2004
      0 references
      sequence
      0 references
      uniform distribution
      0 references
      diaphony
      0 references
      van der Corput sequence
      0 references
      The notion of discrepancy \(D_N(\mathbf x_n)\) was introduced to measure the distribution deviation of the sequence \(\mathbf x_1,\dots,\mathbf x_N\) in \([0,1)^s\) from the expected ideal one [see \textit{H. Niederreiter}, Random number generation and quasi-Monte Carlo methods, CBMS-NSF Regional Conference Series in Applied Mathematics. 63. Philadelphia, PA: SIAM (1992; Zbl 0761.65002)]). The limit \(D_N(\mathbf x_n)\to0\) as \(N\to\infty\) implies uniform distribution of \(\mathbf x_n\), \(n=1,2,\dots\) There are known: extremal discrepancy, star discrepancy, \(L^2\) discrepancy, polynomial discrepancy, Abel discrepancy, isotropic discrepancy, etc., and also diaphony (a survey of discrepancies can be found in the book of the reviewer and \textit{Š. Porubský} [Distribution of sequences. A sampler. Schriftenreihe der Slowakischen Akademie der Wissenschaften 1. Bern: Peter Lang. (2005; Zbl 1078.11051)]. The concept of diaphony was introduced by \textit{P. Zinterhof} [Österr. Akad. Wiss., Math.-Naturw. Kl., S.-Ber., Abt. II, 185, 121--132 (1976; Zbl 0356.65007)]. Replacing trigonometric functions by Walsh functions it is transformed to dyadic diaphony by \textit{P. Hellekalek} and \textit{H. Leeb} [Acta Arith. 80, No. 2, 187--196 (1997; Zbl 0868.11034)]. In this paper, using Chrestenson functions in base \(b\), it is further generalized to the \(b\)-adic diaphony \(F_N(\mathbf x_n)\). For the sequence \(\mathbf x_1=(x_{1,1},\dots,x_{1,s}),\dots,\mathbf x_N=(x_{N,1},\dots,x_{N,s})\) whose coordinates of all points are \(b\)-adic rationals, the authors give the expression \[ (F_N(\mathbf x_n))^2=((b+1)^s-1)^{-1}N^{-2}\sum_{m,n=1}^N\phi(\mathbf x_m\dot{-}\mathbf x_n), \] where \(\mathbf x_m\dot{-}\mathbf x_n=(x_{m,1}\dot{-}x_{n,1},\dots,x_{m,s}\dot{-}x_{n,s})\), and if \(x=0.x_1x_2\dots x_k\), \(y=0.y_1y_2\dots y_k\), \(z=0.z_1z_2\dots z_k\) are \(b\)-adic expansions, then \(x\dot{-}y=z\) if \(x_i-y_i\equiv z_i\pmod b\). Here \(\phi(\mathbf x)=-1+\prod_{j=1}^s\varphi(x_j)\) for \(\mathbf x=(x_1,\dots,x_s)\), \(\varphi(x)=(b+1)-(b+1)b^{1+[\log_b x]}\) for \(x\in(0,1)\) and \(\varphi(0)=b+1\). Using this the authors obtain the exact order \(F_N(x_n)=O(\sqrt{\log N}/N)\) for one-dimensional generalized van de Corput sequence \(x_n\), \(n=1,2,\dots\), introduced by \textit{H. Faure} [Bull. Soc. Math. France 109, 143--182 (1981; Zbl 0488.10052)] as follows: If \(n=a_ka_{k-1}\dots a_1a_0\) is \(b\)-adic digit expansion of \(n\) and \(\sigma_j\), \(j=0,1,\dots\) is a sequence of permutations of digits \(\{0,1,\dots,b-1\}\), then \(x_n=0.\sigma_0(a_0)\sigma_1(a_1)\dots\sigma_k(a_k)\).
      0 references

      Identifiers