Representation of numbers by the norm forms with two dominating variables (Q1216334)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3475526
Language Label Description Also known as
default for all languages
No label defined
    English
    Representation of numbers by the norm forms with two dominating variables
    scientific article; zbMATH DE number 3475526

      Statements

      Representation of numbers by the norm forms with two dominating variables (English)
      0 references
      1974
      0 references
      Let \(K\) be an algebraic number field of degree \(n\ge 3\) over \(\mathbb Q\), \(\lambda\in K\), \(\vert\overline{\lambda}\vert\) (house) the maximum of the absolute values of all conjugates \(\lambda^{(i)}\) \((i=1, 2, \ldots,n)\), \(\mathrm{Norm}(\lambda)\) the absolute norm of \(\lambda\). In this note the author considers the following generalization of the equation of Thue \[ \mathrm{Norm}(x+ \alpha y+ \lambda = A, \tag{1} \] with \(\alpha\) an algebraic integer of degree \(n\) from \(K\), \(A\ne 0\) a rational integer, \(x\) and \(y\) rational integral variables, \(\lambda\in K\), an algebraic integral variable and where \[ \vert\overline{\lambda}\vert \le X^{1 - \delta},\quad X = \max(\vert x\vert, \vert y\vert) \text{ with a small positive }\delta. \tag{2} \] Theorem 1. There are effectively computable values \(c_1 > 0\) and \(c_2 > 0\), depending on \(n\) and the regulator of \(K\), such that for any \(\delta\) in the interval \(0 < \delta\le c_1\) \((c_1 < 1)\) from (1) and (2) follows \[ X < (nH\vert A\vert)^{c_2(1/\delta)\log(1/\delta)}, \] where \(H\) is the height of \(\alpha\). Theorem 2. Let \(K\) be a real algebraic number field of degree \(n\ge 3\), and let \(1, \alpha_1, \ldots, \alpha_{n-1}\) be an integral basis of \(K\) (which is supposed to be \(\mathbb Q(\alpha_1)\). Then \(c_3>0\), \(c_4>0\) and \(X_0 >0\) can be determined effectively through \(\alpha_1, \ldots, \alpha_{n-1}\) such that for any \(\delta\) in the interval \(0 < \delta\le c_3\) \((c_3 < 1)\)) and any \(X>X_0\) the inequalities \[ \Vert \alpha_1x_1 + \cdots +\alpha_{n-1}x_{n-1}\Vert \le X^{-n+1+c_4\delta/\log(1/\delta)} \] and \(\vert x_1\vert \le X\), \(\max \vert x_1\vert \le X^{1-\delta}\) \((,i =2,3, \ldots,n - 1)\) have no solutions in rational integers \((x_1, \ldots, x_n) \ne (0)\) \((\Vert \cdots\Vert\) denotes the distance to the nearest integer). The proofs use some results proved by the author in [Math. USSR, Izv. 5(1971), 1003--1019 (1972); translation from Izv. Akad. Nauk SSSR, Ser. Mat. 35, 991--1007 (1971; Zbl 0231.10022), and Math. USSR, Izv. 6(1972), 705--734 (1973); translation from Izv. Akad. Nauk SSSR, Ser. Mat. 36, 712--741 (1972; Zbl 0244.10030)].
      0 references
      generalization of Thue equation
      0 references

      Identifiers