A generalization of a theorem about gapsets with depth at most 3 (Q6117069): Difference between revisions

Let \(\mathbb{N}\) denote the set of nonnegative integers, and let \(S\) be a numerical semigroup, that is, a submonoid of \((\mathbb{N},+)\) with finite complement in \(\mathbb{N}\). The cardinality of \(\mathbb{N}\setminus S\) is called the genus of \(S\), and the smallest positive integer belonging to the semigroup is known as its multiplicity. Let \(m\) be the multiplicity of the numerical semigroup \(S\). For \(i\in\{0,\dots,m-1\}\), let \(w_i\) be the smallest integer in \(S\) congruent with \(i\) modulo \(m\); write \(w_i=mk_i+i\). Then \((k_1,\dots,k_{m-1})\) are called the Kunz coordinates of \(S\) (we allways drop \(k_0\) because it is equal to zero). Let \(n'_{g,l}\) be the set of numerical semigroups with genus \(g\) such that all its Kunz coordinates are between \(l\) and \(2l+1\). The authors prove that \[ \sum_{i=l}^{2l}n'_{g-i,l}\le n'_{g,l}\le \sum_{i=l}^{2l+1}n'_{g-i,l}. \] For \(l=1\), one easily derives Theorem 5.4 in [\textit{S. Eliahou} and \textit{J. Fromentin}, J. Comb. Theory, Ser. A 169, Article ID 105129, 19 p. (2020; Zbl 1456.20064)].