Carleson-type embeddings with closed range

DOI10.1007/S00209-025-03881-WMaRDI QIDQ6898202zbMATH OpenFDO

Authors Konstantin M. Dyakonov

Publication date 18 November 2025

Published in Mathematische Zeitschrift (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://doi.org/10.1007/S00209-025-03881-W

zbMATH Keywords

Bergman spaces Carleson measure closed range Hardy spaces interpolation sampling

Mathematics Subject Classification ID

Hardy spaces (30H10) Bergman spaces and Fock spaces (30H20) Operators on complex function spaces (47B91)

Review text

The paper under review deals with characterizing the Carleson measures \(\mu\) on the open unit disk \(\mathbb{D}=\{z\in\mathbb{C}: |z|<1\}\) for which the image of the Hardy (resp. Bergman) space \(H^p\) (resp. \(A^p\)) in \(L^p(\mu)\) is closed. \par Let \(\mathbb{T}\) denote the boundary of the unit disk \(\mathbb{D}\). For \(0<p<\infty\), the Hardy space \(H^p\) consists of analytic functions on the open unit disk for which \N\[\N\Vert f\Vert^p_{H^p}=\sup_{0<r<1}\int_{\mathbb{T}}|f(r\zeta)|^p dm(\zeta)<\infty,\N\]\Nwhere \(m\) is the normalized arc-length measure on the unit circle. Let \(\mu\) be a positive Borel measure on the unit disk, and let \(L^p(\mu)\) denote the space of Borel measurable functions \(f:\mathbb{D}\to \mathbb{C}\) satisfying \N\[\N\Vert f\Vert^p_{L^p}=\int_{\mathbb{D}}|f(z)|^p d\mu(z)<\infty.\N\]\NThe measure \(\mu\) is said to be a Carleson measure provided that for some \(0<p<\infty\) (and indeed, for every \(p\)) we have \(H^p\subset L^p(\mu)\); meaning that the embedding (or inclusion) map \N\[\Nj_p(\mu):H^p\hookrightarrow L^p(\mu)\N\]\Nis bounded. The paper under review deals with the question for which Carleson measures \(\mu\) the above embedding operator has closed range in \(L^p(\mu)\). Similarly, \(\mu\) is called a \((p,q)\)-Carleson measure if \N\[\Nj_{p,q}(\mu):H^p\hookrightarrow L^q(\mu)\N\]\Nis a bounded operator. Now, the above question takes the following form: for which \((p,q)\)-Carleson measures \(\mu\), the operator \(j_{p,q}\) has closed range in \(L^q(\mu)\). Note that if \(\mu\) is supported on a finite subset of points in the unit disk, then the inclusion map is surjective and hence its range is closed. The first main result of the paper under review is:\N\NTheorem 1. Let \(0<p,q<\infty\), and let \(\mu\) be a \((p,q)\)-Carleson measure supported on an infinite subset of the unit disk. Then the following are equivalent:\N\N(i)~ The range of \(j_{p,q}\) is closed in \(L^q(\mu)\);\N\N(ii)~\(p=q\) and \(\mu=\sum_{n} a_n\delta_{z_n}\) where the sequence \(\{z_n\}\subset \mathbb{D}\) satisfies \N\[\N\inf_k\prod_{j:j\neq k}\Big |\frac{z_j-z_k}{1-\overline{z_j}z_k}\Big |>0,\N\]\Nwhere \(\delta_{z_n}\) denotes the unit point mass at \(z_n\), and the positive numbers \(a_n\) satisfy \(a_n\asymp 1-|z_n|\).\N\NIn the last part of the paper, the author considers the same problem for the Bergman space \(A^p\) consisting of analytic functions \(f\) on the unit disk for which \N\[\N\Vert f\Vert^p_{A^p}=\int_{\mathbb{D}}|f(z)|^p d\sigma(z)<\infty,\N\]\Nwhere \(\sigma\) denotes the normalized area measure in the unit disk. Likewise, a Borel measure \(\mu\) on the unit disk is called a Carleson measure for \(A^p\) (or a Bergman-Carleson measure) if \(A^p\subset L^p(\mu)\). In the same way, \(\mu\) is called a \((p,q)\)-Bergman-Carleson measure if \(A^p\subset L^q(\mu)\). To answer the same question for \N\[\Nj_{p,q}(\mu):A^p\hookrightarrow L^q(\mu)\N\]\Nwe need to recall that a Bergman-Carleson measure \(\mu\) for \(A^p\) is called a sampling measure for \(A^p\) provided that \N\[\N\Vert f\Vert_{A^p}\asymp \Vert f\Vert_{L^p(\mu)},\N\]\Nhere the notation \(A\asymp B\) (for a nonnegative quantity \(A\) and a positive quantity \(B\)) means that there is a number \(c\ge 1\) such that \N\[\N\frac{1}{c}B\le A\le cB.\N\]\NA sequence \(\{z_k\}\) of distinct points in the unit disk is called an interpolating sequence for \(A^p\) if the interpolation problem \(f(z_k)=w_k\) for every positive integer \(k\) has a solution \(f\in A^p\) whenever \N\[\N\sum_{k=1}^\infty (1-|z_k|^2)^2 |w_k|^p<\infty.\N\]\NThe second main result of the paper under review reads as follows:\N\NTheorem 2. Let \(0<p,q<\infty\), and let \(\mu\) be a \((p,q)\)-Bergman-Carleson measure supported on an infinite subset of the unit disk. The following are equivalent:\N\N(i)~ The range of \(j_{p,q}\) is closed in \(L^q(\mu)\);\N\N(ii)~\(p=q\) and one of the two conditions hold: either \(\mu\) is a sampling measure for \(A^p\), or \(\mu=\sum_{n} a_n\delta_{z_n}\) where \(\{z_n\}\subset \mathbb{D}\) is an interpolating sequence for \(A^p\) and the positive numbers \(a_n\) satisfy \(a_n\asymp (1-|z_n|)^2\).

Cites work

This page was built for publication: Carleson-type embeddings with closed range

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6898202)