An upper bound on the number of classes of perfect unary forms in totally real number fields (Q6591606)

From MaRDI portal





scientific article; zbMATH DE number 7900421
Language Label Description Also known as
English
An upper bound on the number of classes of perfect unary forms in totally real number fields
scientific article; zbMATH DE number 7900421

    Statements

    An upper bound on the number of classes of perfect unary forms in totally real number fields (English)
    0 references
    0 references
    0 references
    22 August 2024
    0 references
    Let \(K\) be a totally real number field of degree \(r\) and let \(\omega_1, \ldots, \omega_r\) be a basis in the ring of algebraic integers \(\mathcal{O}_K\) in \(K\). For each totally positive \(a\in K\) (denoted by \(a\in K_{>>0}\)) one can associate the positive definite rational quadratic form\N\[\N\mathrm{Tr}_{K/\mathbb{Q}} ax^2 = \mathrm{Tr}_{K/\mathbb{Q}} a(x_1\omega_1+\ldots + x_r\omega_r)^2\N\]\Nof \(m\) variables. Denote by \(\mu(a)=\min \{\mathrm{Tr}_{K/\mathbb{Q}} av^2 \ | \ 0\ne v\in \mathcal{O}_K \}\) and \(\mathcal{M}(a)= \{v\in \mathcal{O}_K \ | \ \mathrm{Tr}_{K/\mathbb{Q}} av^2 = \mu(a)\}\) the minimum and the set of minimal vectors respectively. A unary quadratic form \(ax^2\) (\(a\in K\) is totally positive) is called perfect, if the system of linear equations \(\mathrm{Tr}_{K/\mathbb{Q}} av^2 = \mu(a)\), \(v\in \mathcal{M}(a)\), yields a unique solution \(a\) (see [\textit{M. Koecher}, Math. Ann. 141, 384--432 (1960; Zbl 0095.25301)]).\N\NUsing the volumetric argument (see [\textit{W. P. J. van Woerden}, Adv. Math. 365, Article ID 107031, 12 p. (2020; Zbl 1444.11150)]) the authors obtained an upper bound for the number of unary perfect forms (up to equivalence and scaling) (see Theorem 1). This upper bound depends on the covering radius of the log unit lattice in the infinity norm and additive Hermite-Humbert number\N\[\N\gamma_K=\sup_{a\in K_{>>0} } \frac{\mu(a)}{\mathrm{Nm}_{K/\mathbb{Q}}(a)^{1/[K\colon \mathbb{Q}]}}.\N\]\NMoreover, the obtained upper bound was simplified under the assumption that \(K\) is a Galois extension of prime degree (Theorem 2) or \(K\) is a maximal totally real subfield of a cyclotomic field (Theorem 3). In both cases neither covering radius nor \(\gamma_K\) were used.
    0 references
    0 references
    lattices
    0 references
    quadratic forms
    0 references
    perfect forms
    0 references
    number fields
    0 references
    trace forms
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references