Inverse problems for linear forms over finite sets of integers

From MaRDI portal
Publication:3597602

zbMATH Open1163.11067arXiv0708.2304MaRDI QIDQ3597602FDOQ3597602


Authors: Melvyn B. Nathanson Edit this on Wikidata


Publication date: 9 February 2009

Abstract: Let f(x_1,x_2,...,x_m) = u_1x_1+u_2 x_2+... + u_mx_m be a linear form with positive integer coefficients, and let N_f(k) = min{|f(A)| : A subseteq Z and |A|=k}. A minimizing k-set for f is a set A such that |A|=k and |f(A)| = N_f(k). A finite sequence (u_1, u_2,...,u_m) of positive integers is called complete if {sum_{jin J} u_j : J subseteq {1,2,..,m}} = {0,1,2,..., U}, where U=sumj=1muj. It is proved that if f is an m-ary linear form whose coefficient sequence (u_1,...,u_m) is complete, then N_f(k) = Uk-U+1 and the minimizing k-sets are precisely the arithmetic progressions of length k. Other extremal results on linear forms over finite sets of integers are obtained.


Full work available at URL: https://arxiv.org/abs/0708.2304




Recommendations





Cited In (13)





This page was built for publication: Inverse problems for linear forms over finite sets of integers

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3597602)