Allgemeine Kongruenzklasseneinteilungen der Ideale einfacher Algebren über algebraischen Zahlkörper und ihre L-Reihen. (Q5771379)
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 3031684
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Allgemeine Kongruenzklasseneinteilungen der Ideale einfacher Algebren über algebraischen Zahlkörper und ihre L-Reihen. |
scientific article; zbMATH DE number 3031684 |
Statements
Allgemeine Kongruenzklasseneinteilungen der Ideale einfacher Algebren über algebraischen Zahlkörper und ihre L-Reihen. (English)
0 references
1938
0 references
0.8437618613243103
0 references
0.7392174601554871
0 references
0.7370745539665222
0 references
0.7341848015785217
0 references
Let \(\mathfrak A = \mathfrak A/K\) be a normal simple algebra, of degree \(n\) over the algebraic number field \(K\), which is not a total definite quaternion algebra over \(K\), that is, if \(\mathfrak u\) denotes the product of the infinite prime spots of \(K\) which are ramified in \(\mathfrak A\), then either \(n > 2\) or \(\mathfrak u\) does not contain all infinite prime spots of \(K\) [cf. the author, Math. Z. 43, 481--494 (1938; Zbl 0018.202)]. First, the author studies congruence classes of the ideals of \(\mathfrak A\), analogous to the ray classes etc. in the class field theory, and obtains an equivalence criterion in terms of norms which is a generalization of that in his earlier paper (loc. cit.). Second, he proves the following ``norm theorem for arithmetical progressions'': If \(\mathfrak J_1\) is a maximal order of \(\mathfrak A\), \(\mathfrak F_{11}\) an integral two-sided \(\mathfrak J_1\)-ideal, and if \(\alpha\) and \(b\) are numbers in \(\mathfrak J_1\) and \(K\), resp., such that \[ b\equiv n(\alpha) \mod^{.} \,\mathfrak F_{11}, \quad b\equiv n(\alpha)\equiv 1\pmod{\mathfrak u}, \] then \(\mathfrak J_1\) contains a number \(\beta\) such that \[ \beta \equiv \alpha \pmod{\mathfrak F_{11}}, \quad, n(\beta) = b. \] The dot denotes multiplicative congruence (as below) and \(n(\;)\) is the reduced norm relative to \(K\). This theorem yields as corollaries the Hasse-Schilling norm theorem [Zbl 0014.006)]; the known equivalence criteria for the ideals of \(\mathfrak A\), including those mentioned above; and the theorem that the Euclidean algorithm holds in every maximal order of \(\mathfrak A\) if it holds in \(K \bmod \mathfrak u\). The converse of this last theorem is conjectured to be false. Finally, the author obtains two functional equations for the \(L\)-series of the congruence classes. These yield conductor formulas, an identity for hypercomplex Gaussian sums, and the result that the zeta-functions of the classes of the same congruence classification have the same residue at \(s = 1\), which gives information concerning the unit groups of different maximal orders. To define the congruence classes, the author proceeds as follows. Each of the totality of ideals \(\mathfrak F_{11}, \mathfrak F_{22}, \ldots,\) with maximal orders \(\mathfrak J_1, \mathfrak J_2, \ldots,\) and the same norm \(n(\mathfrak F_{11}) = n(\mathfrak F_{11}) = \cdots\), is uniquely determined by \(\mathfrak J_\nu\) and the norm. Let \(\mathfrak F\) denote this ideal system, writing \(\mathfrak F_{\nu\nu} = \mathfrak J_\nu \mathfrak F = \mathfrak F\mathfrak J_\nu\), \(n(\mathfrak F) = n(\mathfrak F_{\nu\nu}\), etc. If \(\mathfrak F\) is an integral ideal system, an ideal \(\mathfrak N_{12}\) is said to be prime to \(\mathfrak F\) if \(\mathfrak N_{12}^{\mathfrak p} = \mathfrak J_1^{\mathfrak p}\) \((\mathfrak p\)-components) for each prime ideal \(\mathfrak p\) of \(K\) which divides \(n(\mathfrak F)\). The totality of ideals of \(\mathfrak A\) prime to \(\mathfrak F\) and having the same \(\mathfrak p\)-components for all such \(\mathfrak p\) is a subgroupoid \(\mathfrak A_{\mathfrak F}\) of the Brandt groupoid of ideals of \(\mathfrak A\). If \(\alpha\) is an element of \(\mathfrak A\), write \(\alpha \equiv 1 (\mod^{.} \, \mathfrak J_1\mathfrak F)\) (multiplicative congruence) in case there exist \(\beta\) and \(\gamma\) in \(\mathfrak J_1\), prime to \(\mathfrak F\) and such that \(\alpha = \beta^{-1}\gamma\), \(\beta\equiv \gamma\pmod{\mathfrak J_1\mathfrak F}\). Finally two ideals \(\mathfrak N_{i_1i_2}\) and \(\mathfrak N_{i_2i_1}\) are said to be equivalent \(\mod \mathfrak F\) if \[ \sigma_{i_1}\mathfrak N_{i_1i_2}\sigma_{i_2} = \sigma_{i_2} \mathfrak N_{i_2i_1} \sigma_{i_1}, \] with \(\sigma_{i_\nu} \equiv 1 (\bmod^{.}\mathfrak J_{i_\nu}\mathfrak F)\), \(\nu= 1,\ldots,4\). If further \(\mathfrak J_{i_1} = \mathfrak J_{i_2}\), they are said to be left equivalent \(\bmod \mathfrak F\). The classes so obtained are the ray classes, left ray classes, resp., \(\bmod \mathfrak F\). The equivalence criteria etc. are obtained from the theorem: If \(\mathfrak F\) is an integral ideal system in \(\mathfrak A\) then \(\mathfrak F\) determines a divisor \(\mathfrak f\) in \(K\) \((\mathfrak f\) is a factor of \(\mathfrak u\, n(\mathfrak F))\) such that an ideal in \(\mathfrak A\) is principal \(\bmod \mathfrak F\) if and only its norm is principal \(\bmod \mathfrak f\). The ray classes in a subgroupoid \(\mathfrak A_{\mathfrak F}\) form a ray class groupoid \(\Sigma_{\mathfrak F}\). To obtain the more general congruence classes, the author employ lemmas on groupoids, the mapping \(\mathfrak N_{ij} \to n(\mathfrak N_{ij})\) of \(\Sigma_{\mathfrak F}\) on \(\sigma_{\mathfrak f}\), where \(\sigma_{\mathfrak f}\) is the ray dass group \(\bmod \mathfrak f\) in \(K\), and homomorphisms \(\sigma_{\mathfrak f} \sim \pi_{\mathfrak f}\), where if \(\pi_{\mathfrak f}\) is an extended class group in \(K\). A corresponding norm criterion for the extended classes is obtained.
0 references