Planar Sidonicity and quasi-independence for multiplicative subgroups of the roots of unity (Q2373554)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Planar Sidonicity and quasi-independence for multiplicative subgroups of the roots of unity |
scientific article |
Statements
Planar Sidonicity and quasi-independence for multiplicative subgroups of the roots of unity (English)
0 references
12 July 2007
0 references
In this interesting paper, the authors investigate Sidon and quasi-independent properties for subsets of the roots of unity, viewed as subsets of the discrete complex plane \(\mathbb C\). Recall that a subset \(Q\) of an abelian group \(G\) (the law noted additively) is quasi-independent (resp. independent) if for every \(\varepsilon_q\in\{-1,0,1\}\) (resp. \(\varepsilon_q\in\mathbb Z\)), where almost every \(\varepsilon_q=0\), we have \(\displaystyle\sum_{q\in Q}\varepsilon_qq=0\) only when every \(\varepsilon_q=0\). A subset \(S\) of an abelian group \(G\) is a Sidon set if every bounded complex-valued function on \(S\) is the restriction of the Fourier-Stieltjes transform of a Borel measure on \(G\). This is actually the dual definition of the classical one: see the survey of \textit{M. Dechamps-Gondim} [Publ. Math. Orsay 84--01, Exp. No.~7, 51 pp. (1984; Zbl 0537.43018)] to know more on the subject until the mid-eighties. Sidon sets and, more generally lacunary sets belong to an old field of harmonic analysis. This was renewed in the early 1970's after the results of Drury (on the union of Sidon sets) and Rider, then later with the (deep) results of Pisier (and then Bourgain), providing some arithmetical characterizations of these sets. In the paper under review, several results are obtained in the specific framework of the set of the roots of unity. The authors study the size of the largest quasi-independent subset of the group of the \(n\)th roots of unity (see Theorem 4.1 and Section 4.). We are not going to list them all; the more so since the paper begins with a nice overview. Nevertheless, let us point out two theorems. First, Theorem 1.4 states: Let \(P\) be a set of primes, and \(W\) be the multiplicative subset of \(\mathbb Z\) generated by \(P\). Then \(E=\{e^{2ia/m}: a\in\mathbb Z\;\text{and}\;m\in W\}\) is a finite union of independent sets iff \(\displaystyle\sum_P\tfrac{1}{p}<\infty\). Theorem 3.1. shows that Pisier's criterion can be weakened in the particular framework of the paper under review: it suffices to test fewer finite sets for the proportionality property (in Pisier's theorem): A subset \(A\) of the set of the roots of unity is Sidon iff there is an integer \(k\) such that for every positive, squarefree integer \(n\geq 1\), for every coset \(U\) of the group of the \(n\)th roots of unity and every finite \(F\subset A\cap U\), there is some quasi-independent set \(E\subset F\) such that \(\text{card}(E)\geq\frac{1}{k}\text{card}(F)\).
0 references
Sidon property
0 references
independent sets
0 references
roots of unity
0 references