English

Maximum Gap in (Inverse) Cyclotomic Polynomial

Number Theory 2011-01-25 v1

Abstract

Let g(f)g(f) denote the maximum of the differences (gaps) between two consecutive exponents occurring in a polynomial ff. Let Φn\Phi_n denote the nn-th cyclotomic polynomial and let Ψn\Psi_n denote the nn-th inverse cyclotomic polynomial. In this note, we study g(Φn)g(\Phi_n) and g(Ψn)g(\Psi_n) where nn is a product of odd primes, say p1<p2<p3p_1 < p_2 < p_3, etc. It is trivial to determine g(Φp1)g(\Phi_{p_1}), g(Ψp1)g(\Psi_{p_1}) and g(Ψp1p2)g(\Psi_{p_1p_2}). Hence the simplest non-trivial cases are g(Φp1p2)g(\Phi_{p_1p_2}) and g(Ψp1p2p3)g(\Psi_{p_1p_2p_3}). We provide an exact expression for g(Φp1p2).g(\Phi_{p_1p_2}). We also provide an exact expression for g(Ψp1p2p3)g(\Psi_{p_1p_2p_3}) under a mild condition. The condition is almost always satisfied (only finite exceptions for each p1p_1). We also provide a lower bound and an upper bound for g(Ψp1p2p3)g(\Psi_{p_1p_2p_3}).

Keywords

Cite

@article{arxiv.1101.4255,
  title  = {Maximum Gap in (Inverse) Cyclotomic Polynomial},
  author = {Hoon Hong and Eunjeong Lee and Hyang-Sook Lee and Cheol-Min Park},
  journal= {arXiv preprint arXiv:1101.4255},
  year   = {2011}
}
R2 v1 2026-06-21T17:15:17.246Z