|
This article is cited in 2 scientific papers (total in 2 papers)
Uniform distribution of non-divisible vectors in an integer space
V. I. Arnol'd Steklov Mathematical Institute, Russian Academy of Sciences
Abstract:
A vector in an integer space is said to be divisible if it is
the product of another vector in this space and an integer exceeding 1.
The uniform distribution of a set of integer vectors means that the number
of points of this set in the image of a domain in $n$-dimensional space
under $N$-fold dilation is asymptotically proportional
to the product of $N^n$ and the volume of the domain as $N\to\infty$.
The constant of proportionality (called the density of the set)
is equal to $1/\zeta(n)$ for the set of non-divisible vectors in
$n$-dimensional integer space (where $n>1$). For example,
the density of the set of non-divisible vectors on the plane is equal
to $1/\zeta(2)=6/\pi^2\approx 2/3$. It was this discovery that led Euler
to the definition of the zeta-function.
The proof of the uniform distribution of the set of non-divisible integer
vectors is published here because there are arbitrarily large domains
containing no non-divisible vectors.
We shall show that such domains are situated only far from the origin
and are infrequent even there. Their distribution is also uniform
and has a peculiar fractal character, which has not yet been studied
even at the empirical computer-guided level or even for $n=2$.
Keywords:
crystal lattice, zeta-function, prime numbers, trigonometric sums, inclusion/exclusion, density of distribution, Legendre–Chebyshev theorem.
Received: 21.02.2008
Citation:
V. I. Arnol'd, “Uniform distribution of non-divisible vectors in an integer space”, Izv. Math., 73:1 (2009), 21–29
Linking options:
https://www.mathnet.ru/eng/im2773https://doi.org/10.1070/IM2009v073n01ABEH002436 https://www.mathnet.ru/eng/im/v73/i1/p21
|
|