Dirichlet's problem with entire data posed on an ellipsoidal cylinder (Q507185): Difference between revisions

Consider the Dirichlet problem on a domain \(\Omega \) in \(\mathbb{R}^{n}\) with boundary data \(f\). In the special case where \(\Omega \) is an ellipsoid, the first author and \textit{H. S. Shapiro} showed that an entire boundary function yields an entire solution [Bull. Lond. Math. Soc. 24, No. 5, 456--468 (1992; Zbl 0799.35030)]. Further in [Comput. Methods Funct. Theory 10, No. 2, 519--554 (2010; Zbl 1207.35013)], the third author has gone a long way to verify that ellipsoids are the only bounded domains for which this conclusion may be drawn. Now consider unbounded domains of the form \(\Omega =D\times \mathbb{R}\), where \(D\) is an ellipsoid in \(\mathbb{R}^{n-1}\). If \(f\) is an entire function on \(\mathbb{C}^{n}\) with order \(\rho (f)<1\), then the Poisson integral of \(f\) in \(\Omega \) exists and solves the Dirichlet problem on \(\Omega \). The authors show that this solution again extends to a harmonic function on all of \(\mathbb{R}^{n}\) (and an entire function on \(\mathbb{C}^{n}\)). They conjecture that, for arbitrary entire boundary data, every solution to the Dirichlet problem on an ellipsoidal cylinder \(\Omega \) is entire.