Closed and prime ideals of weak Bezout type in \(H^ \infty\) (Q1181452): Difference between revisions

One of the hard problems in the theory of Banach algebras of analytic functions is the characterization of ideals. In the disk algebra this problem was solved by Beurling and Rudin. An ideal \(I\subset H^ \infty\) (\(H^ \infty\) is the algebra of bounded analytic functions in the open unit disk) is said to have weak Bezout property if \(I\) contains the greatest common divisor of every pair of elements in \(I\). The goal of the Mortini paper is to study the ideals in \(H^ \infty\) with weak Bezout property. In the ring \(H(G)\) of all analytic functions in a planar domain \(G\) every ideal has the weak Bezout property. In the algebra \(H^ \infty\) this no longer holds as the example \(I=(1-z,\exp^{-(1+z)/(1- z)})\) shows. Then the author asks if the weak Bezout prime ideals which contain a Blaschke product are in a one-to-one correspondence with some special filters of zero sets (or some special filters of measures). Surprising was the following result: let \(P\) be a prime ideal which is contained in a maximal ideal of the Shilov boundary of \(H^ \infty\). Then \(P\) has the weak Bezout property. Also the author proves that \(I\) (\(I\) is a closed ideal in \(H^ \infty\)) has the weak Bezout property if and only if there exists an inner function \(u\) and a closed subset \(E\) of the Shilov boundary of \(H^ \infty\) such that \(I=\{uf\in H^ \infty\): \(\widehat {f}|_ E=0\}\). In section 3 the author notes that every maximal ideal which does not belong to the Shilov boundary and which is different from \(M(z_ 0)\) (\(z_ 0\in\mathbb{D})\) contains an infinite chain of prime ideals. Then he asks, if one can say more: Does every prime ideal \(P\neq M(z_ 0)\) contain an infinite chain of prime ideals? Some interesting examples are given in section 3.