R R -boundedness of Poisson operators
Robert Denk, Nick Lindemulder, Jörg Seiler · 2025
Read R R -boundedness of Poisson operators by Robert Denk, Nick Lindemulder, Jörg Seiler (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-17557.txt.
Introduction
Robert Denk, Nick Lindemulder, and Jorg Seiler study a class of math operators tied to edge problems in a 2025 paper. The operators, named for Poisson, act on a half-space and shift with a free value. The team checks when whole families of them stay bounded in a strong shared sense across several scales of function spaces. That bound is the key step that lets edge problems with live boundary rules gain the best kind of time control.
Catalog record
| CREATOR | Robert Denk, Nick Lindemulder, Jörg Seiler |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 94 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-17557
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
R R -boundedness of Poisson operators Robert Denk, Nick Lindemulder, Jörg Seiler Abstract . We investigate the R R -boundedness of parameter-dependent families of Poisson operators on the half-space ℝ + n \mathbb{R}^{n}_{+} in various scales of function spaces. Applications concern maximal L q L_{q} -regularity for boundary value problems with dynamic boundary conditions. 2020 Mathematics Subject Classification 35J40 (primary); 46E35, 47D06, 35K35 (secondary) 1. Introduction Poisson operators arise as solution operators to boundary value problems with inhomogeneous boundary conditions. A simple example is the parameter-dependent heat equation in the half-space \mathbb{R}^{n}_{+}:=\{x\in\mathbb{R}^{n}:x_{n}>0\} , (1-1) \displaystyle\mu^{2}u-\Delta u \displaystyle=f \displaystyle\text{ in }\mathbb{R}^{n}_{+}, \displaystyle\gamma_{0}u \displaystyle=g \displaystyle\text{ on }\mathbb{R}^{n-1}, where \gamma_{0}u=u|_{\mathbb{R}^{n-1}} stands for the boundary trace. If \mu is complex with positive real part (and f and g are suitable) this equation is uniquely solvable and the solution can be written in the form u=R_{\mu}f+K_{\mu}g . The operator R_{\mu} is the resolvent of the Dirichlet-Laplacian, K_{\mu} is an example of a parameter-dependent Poisson operator. In the classical L^{p} -setting, the most natural solution space for the first equation in ( 1-1 ) is H_{p}^{2}(\mathbb{R}^{n}_{+}) , the classical second-order Sobolev (or Bessel potential) space. To obtain a solution in this space, one needs to impose that the boundary datum g belongs to B_{pp}^{2-1/p}(\mathbb{R}^{n-1}) , where B_{pp}^{s} stands for the Besov space or order s . However, there are several applications where the boundary data space has less regularity. For instance, this is the case for stochastic PDEs with boundary noise, where g stands for the derivative of a Brownian motion on \mathbb{R}^{n-1} , which is known to belong to some Besov space of negative regularity. Another example is given by the class of boundary value problems with dynamical boundary conditions, which are typically understood as Cauchy problems in the product space L^{p}(\mathbb{R}^{n}_{+})\times L^{p}(\mathbb{R}^{n-1}) . For considering boundary data of low regularity in ( 1-1 ), one has to take care of the well-posedness of the boundary trace \gamma_{0} . This trace is continuous as a map from H_{p}^{s}(\mathbb{R}^{n}) to B_{pp}^{s-1/p}(\mathbb{R}^{n-1}) if and only if s>1/p (see, e.g., [ JS08 ] ). However, if we modify the first space to H_{p,\Delta}^{s}(\mathbb{R}^{n}_{+}):=\{u\in H_{p}^{s}(\mathbb{R}^{n}_{+})\mid\Delta u\in L^{p}(\mathbb{R}^{n}_{+})\} , the boundary trace \gamma_{0}\colon H_{p,\Delta}^{s}(\mathbb{R}^{n}_{+})\to B_{pp}^{s-1/p}(\mathbb{R}^{n-1}) is continuous for all s\in\mathbb{R} , so the formulation in ( 1-1 ) makes sense. Surprisingly, the solution operator K_{\mu}\colon B_{pp}^{s-1/p}\to H_{p}^{s}(\mathbb{R}^{n}_{+}) is continuous for all s\in\mathbb{R} , as was shown by Grubb and Kokholm in [ GK93 ] . In the same paper, uniform estimates of R_{\mu} and K_{\mu} in parameter-dependent norms were established. However, these estimates are not sufficient for the study of time-dependent problems like (1-2) \displaystyle\partial_{t}u(t)-\Delta u(t) \displaystyle=f(t) \displaystyle\text{ in }(0,\infty)\times\mathbb{R}^{n}_{+}, \displaystyle\gamma_{0}u(t) \displaystyle=g(t) \displaystyle\text{ on }(0,\infty)\times\mathbb{R}^{n-1}, in the setting of maximal L^{q} -regularity , where for f\in L^{q}((0,\infty);H^{s}_{p}(\mathbb{R}^{n}_{+})) and g\in L^{q}((0,\infty);B_{pp}^{s-1/p}(\mathbb{R}^{n-1})) , we seek a solution in the canonical solution space u\in L^{q}((0,\infty);H_{p}^{s}(\mathbb{R}^{n}_{+}))\cap H_{q}^{1}((0,\infty);H_{p}^{s+2}(\mathbb{R}^{n}_{+})). In this context, a stronger property – called R -boundedness – of the families R_{\mu} and K_{\mu} is needed. In fact, it is known since long that R -boundedness is equivalent to maximal L^{q} -regularity for all q\in(1,+\infty) in the sense of well-posedness in L^{q} - L^{p} -Sobolev spaces in time and space. For more information on maximal regularity theory and related subjects we refer the reader to the recent series of text books of Hytönen, van Neerven, Veraar and Weis [ HvNVW16 , HvNVW17 , HvNVW23 ] and the references therein. The R -boundedness of resolvents like R_{\mu} above is well-known and can be obtained by general results on normally elliptic boundary value problems. Generally, such resolvents can be decomposed in the sum of a parameter-dependent pseudodifferential operator and a so-called parameter-dependent singular Green operator (the resolvent construction can be performed for example in the framework of Boutet de Monvel’s algebra, a pseudodifferential calculus for boundary value problems, introduced in [ BdM71 ] and then developed further over many decades, see [ GK93 , Gru96 , SS94 ] for example). The R -boundedness of such operator-families has been shown in very different settings, see for example [ Hum21 , Proposition 3.6] , [ DK07 , Theorem 3.18] , and [ DS15 , Theorem 8] for some specific results. Main applications include fluid mechanics and geometric evolution equations (see, e.g., [ PS16 ] , [ Sai21 ] , [ Shi20 ] ). Recently, R -boundedness of the solution operator was considered for non-local operators in [ AG25 ] . While the above results and applications are connected with the R -boundedness of the resolvent, the R -boundedness of families of Poisson operators has been addressed only very recently; in fact, we are only aware of the paper [ Hum21 ] . In the present work we obtain new results for the R -boundedness of Poisson operators in a variety of function spaces. Not only we consider Poisson operators originating from classical boundary value problems as descried above, but also a wider class of Poisson operators related to the analysis of so-called ”edge-degenerate” differential operators on manifolds with boundary; see for example [ RS82 , Sch91 ] for more details. In both cases parameter-dependent Poisson operators have the form (K_{\mu}g)(x)=\int_{\mathbb{R}^{n-1}}e^{ix^{\prime}\xi^{\prime}}k(\xi^{\prime},\mu,x_{n})\widehat{g}(\xi^{\prime})\,d\xi^{\prime},\qquad x=(x^{\prime},x_{n})\in\mathbb{R}^{n}_{+}, with a specific structure of the so-called symbol-kernel k , see Definition 3.2 and Definition 4.7 , respectively; here \widehat{g} is the Fourier transform of g . In case of scalar operators (acting on complex-valued functions) the classes of Poisson operators studied in this paper include those of [ Hum21 ] ; on the other hand, we do not consider Banach space valued Poisson operators as has been done there. However, it seems reasonable to expect that our techniques can be extended to more general settings. As an application, we discuss maximal L^{q} -regularity for several boundary value problems with dynamic boundary conditions, including a variant of the Cahn–Hilliard equation and a road-field model based on a Kolmogorov–Petrovskii–Pisconov equation. In both cases, we consider the space L^{p}(\mathbb{R}^{n-1}) on the boundary. This space also appears as a natural boundary space in applications to bulk-surface reaction-sorption-diffusion systems as studied in [ AB21 ] , [ AB24 ] . In contrast to our examples, the solution in these papers has regularity H_{p}^{2} in the interior (bulk) and is coupled to the surface concentration on the boundary. This allows a maximal regularity approach in standard spaces, similar to the analysis in [ DPZ08 ] . The paper is organized as follows. In Section 2 , we provide some background on R -boundedness and random sequence spaces. In Section 3 , we define the first classes of Poisson operators with and without paprameter and show first boundedness and R -boundedness results. Our main results can be found in Sections 4 and 5 , where we consider the R -boundedness of parameter-dependent families of Poisson operators in Besov spaces of negative order (Theorems and 4.3 and 4.5 ), in (weak) L^{p} -spaces (Theorem 4.11 ), in Sobolev spaces (Section 4.3 ), and in further anisotropic and weighted spaces (Section 5 ). Finally, we apply the results to boundary value problems with dynamic boundary conditions in Section 6 . Throughout the paper we will use the following notation. We write \xi=(\xi^{\prime},\xi_{n}) for (co-)vectors \xi\in\mathbb{R}^{n} . For functions a:\mathbb{R}^{n}_{\xi}\to{\mathbb{C}} , we shall denote by a(D)