On the annihilating ideal for trace forms. (Q1424573): Difference between revisions

This article considers the nature of the annihilating ideal in \(Z[x]\) of a class of quadratic forms. A key result is the following, which allows the translation of certain problems on annihilating polynomials to corresponding problems on \(2\)-groups: Let \(G\) be a finite group with \(2\)-Sylow subgroup \(G_2\). Then for any element \(\chi\) in the Burnside ring of \(G\), we have that \(\chi\) is in the trace ideal \(T(G)\) of \(G\) if and only if \(\text{res}_{G_2}^{G}(\chi)\) is in the trace ideal \(T(G_2)\) of \(G_2\). Hence the restriction homomorphism induces an injection \(L(G)/T(G) \to L(G_2)/T(G_2)\), where \(L(G)\) denotes the kernel of the total signature homomorphism. The classes of quadratic forms considered in the article are of the form \(M(G,H)\) where \(H\) is a subgroup of \(G\) containing no nontrivial normal subgroup of \(G\). Then \(M(G,H)\) is the set of quadratic forms \(\phi\) such that \(\phi\) is isometric to the trace form of a field extension \(K[x]/(f(x))/K\) whose Galois closure has Galois group \(G\) and such that the action of the Galois group on the roots of \(f(x)\) is equivalent to the action of \(G\) on the left cosets \(G/H\). An example is given of such a class \(M(G,H)\) whose annihilating ideal is not principal, showing this is not the case in general, but a sizable list of examples for which the annihilating ideal is principal is also given.