Rational functions and kronecker modules

From MaRDI portal





Let \(K\) be an algebraically closed field, \(X\) an indeterminate and \(\mathbb{N}\) denote the natural numbers. A function \(h: K\cup \{\infty \}\to \mathbb{N}\cup \{\infty \}\) is called a ``height function. Using height functions one may classify the torsion-free rank one modules over (i) the polynomial ring \(K[X]\) and (ii) the Kronecker algebra \(A=\left[\begin{matrix} K & K^2 \\ 0 & K \end{matrix}\right]\) [see \textit{U. Fixman}, Trans. Am. Math. Soc. 113, 424--453 (1964; Zbl 0192.13105)]. The paper under review investigates torsion-free \(A\)-modules of rank two which are purely simple. Here ``purely simple is as defined for example by \textit{F. Okoh} in [Trans. Am. Math. Soc. 232, 169--186 (1977; Zbl 0324.15007)]. Such a module \(E\) lies in the middle of a short exact sequence \(0\to M\to E\to N\to 0\) where \(M\), \(N\) are torsion-free rank one and \(M\) is finite-dimensional. Theorem 1.8 of the cited paper (loc. cit.) shows that these modules \(E\) are all constructed, up to equivalence of short exact sequences, from three parameters, namely a positive integer m corresponding to the dimension of \(M\), a height function \(h\) corresponding to the isomorphism type of \(N\) and a \(K\)-linear functional \(\alpha: K(X)\to K\) related to the maps of the sequence. Thus we may use \(E=E(m,h,\alpha)\) to denote these modules. Main theorem: if the cardinality of the support of the height function \(h\) equals that of \(K\) and if \(m\) is a positive integer then there is a function \(\alpha\) on \(K(X)\) making \(E(m,h,\alpha)\) purely simple and causing every submodule \(M\) for which \(E(m,h,\alpha)/M\) is torsion-free to have dimension at least \(m\).











This page was built for publication: Rational functions and kronecker modules

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3748386)