Spectral properties of the canonical solution operator to \(\overline{\partial}\) (Q935041): Difference between revisions
From MaRDI portal
Revision as of 13:03, 28 June 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Spectral properties of the canonical solution operator to \(\overline{\partial}\) |
scientific article |
Statements
Spectral properties of the canonical solution operator to \(\overline{\partial}\) (English)
0 references
31 July 2008
0 references
The authors study the spectral properties of the canonical solution operator \(S\) to the \(\overline \partial\) acting on \((0,1)\)-forms, with holomorphic coefficients in \(L^2(d\mu)\), for measure \(\mu\) with the property that the monomials \(z^\alpha\), \(\alpha \in \mathbb N^n\) are orthogonal in \(L^2(d\mu)\). In this case, the canonical solution operator \(S\) has an explicit formula on the orthogonal basis \(z^\alpha d\overline z_j\), i.e. \[ S(z^\alpha d\overline z_j)=\overline z_j z^\alpha - \frac {c_{\alpha -e_j}}{c_\alpha}z^ {\alpha-e_j}, \alpha \in \mathbb N^n, j=1,2,3\dots,n, \] where \(c_\alpha =(\int |z^\alpha|^2d\mu)^{-1}\). The authors introduce good orthogonal decompositions both on the domain and range of \(S\). The matrix representation of \(S\) with respect to these decompositions is diagonalized, and the diagonal elements of this matrix are simple rational functions of the \(c_\alpha\)'s. Using these informations, the authors prove the following theorems. Let \(\Gamma\) be the set of multi-indices \(\alpha\) which might has one (but not more than one) entry equal to \(-1\): in that case we define \(c_\alpha =0\). Theorem 1.~\(S:A^2_{(0,1)}(d\mu)\to L^2(d\mu)\) is bounded if and only if there exists a constant \(C\) such that \[ \frac {c_{\alpha +e_j}}{c_{\alpha +2e_j}}-\frac {c_\alpha}{c_{\alpha+e_j}}<C \] for all multi-indices \(\alpha \in \Gamma\). Theorem 2.~\(S:A^2_{(0,1)}(d\mu)\to L^2(d\mu)\) is compact if and only if \[ \lim_{\alpha}\bigg(\frac {c_{\alpha +e_j}}{c_{\alpha +2e_j}}-\frac {c_\alpha}{c_ {\alpha+e_j}}\bigg)=0 \] for all \(j=1,2,3,\dots ,n\). Theorem 3. Let \(p>0\). Then \(S:A^2_{(0,1)}(d\mu)\to L^2(d\mu)\) is in the Schatton-\(p\)-class \(S_p\) if and only if \[ \sum_{\alpha \in \Gamma}\bigg(\sum_j\frac {c_{\alpha +e_j}}{c_{\alpha +2e_j}}-\frac {c_\alpha}{c_{\alpha+e_j}}\bigg)^{p/2} <\infty. \] Theorem 4. The canonical solution operator \(S\) is in the Hilbert-Schmidt class if and only if \[ \lim_{k\to \infty}\sum_{\alpha \in \mathbb N^n,|\alpha|=k,1\leq j\leq n}\frac {c_\alpha}{c_{\alpha+e_j}} < \infty. \] The setup in this paper covers the following cases as special cases, (1) bounded Reinhardt domains in \(\mathbb C^n\); (2) weighted \(L^2\) spaces with radialsymmetric weights; (3) weighted \(L^2\) spaces with decoupled radial weights. This paper not only unifies the known results, but also has some new generalizations.
0 references
\(\overline \partial\)-equation
0 references
solution
0 references
compact operator
0 references
Schatton-\(p\)-class
0 references
Hilbert-Schmidt class
0 references
0 references
0 references
0 references
0 references
0 references