Harmonic analysis in Dunkl settings

The Anh Bui · 2024

Read Harmonic analysis in Dunkl settings by The Anh Bui (2024) 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-2412-01067.txt.

Introduction

Bui works in a branch of harmonic analysis built around the Dunkl Laplacian, an operator tied to root systems and a weight function, in this 2024 paper. He studies two families of function spaces attached to that operator, the Besov and Triebel-Lizorkin spaces.

His first result shows these spaces match the same spaces defined on a space of homogeneous type, a setting with a measure that plays well with distance. He then proves multiplier theorems, including a version of the classic Hörmander theorem, for operators built from the Dunkl transform. The findings are new even for Hardy spaces. The work gives analysts a firmer base for studying equations in this weighted setting.

Catalog record

CREATORThe Anh Bui
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE186 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-01067

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Harmonic analysis in Dunkl settings The Anh Bui Abstract . Let L L be the Dunkl Laplacian on the Euclidean space ℝ N {\mathbb{R}^{N}} associated with a normalized root R R and a multiplicity function k ⁡ ( ν ) ≥ 0 , ν ∈ R k(\nu)\geq 0,\nu\in R . In this paper, we first prove that the Besov and Triebel-Lizorkin spaces associated with the Dunkl Laplacian L L are identical to the Besov and Triebel-Lizorkin spaces defined in the space of homogeneous type ( ℝ N , ∥ ⋅ ∥ , d w ) ({\mathbb{R}^{N}},\|\cdot\|,dw) , where d ​ w ​ ( x ) = ∏ ν ∈ R ⟨ ν , x ⟩ k ⁡ ( ν ) ​ d ​ x dw({\rm x})=\prod_{\nu\in R}\langle\nu,{\rm x}\rangle^{k(\nu)}d{\rm x} . Next, consider the Dunkl transform denoted by ℱ \mathcal{F} . We introduce the multiplier operator T m T_{m} , defined as T m ​ f = ℱ − 1 ​ ( m ​ ℱ ​ f ) T_{m}f=\mathcal{F}^{-1}(m\mathcal{F}f) , where m m is a bounded function defined on ℝ N \mathbb{R}^{N} . Our second aim is to prove multiplier theorems, including the Hörmander multiplier theorem, for T m T_{m} on the Besov and Tribel-Lizorkin spaces in the space of homogeneous type ( ℝ N , ∥ ⋅ ∥ , d w ) ({\mathbb{R}^{N}},\|\cdot\|,dw) . Importantly, our findings present novel results, even in the specific case of the Hardy spaces. 2010 Mathematics Subject Classification 42B35, 42B15, 33C52, 35K08 † † authors: 1. Introduction The Dunkl theory is an interesting subject in analysis and can be view as a generalization of Fourier analysis in the Euclidean setting. This foundation can be tracked back by the influential paper of Dunkl [ 5 , 15 ] . Since then, significant progress has been made in its development. See for example [ 14 , 16 , 17 , 33 , 38 , 39 , 40 , 50 , 51 , 34 ] and the references therein. For further information, we refer the reader to [ 41 , 42 ] for more information and references. Motivated by the extensive research conducted in the Dunkl settings, in this paper we will develop the harmonic analysis on the Dunkl setting. Firstly we would like to introduce the framework of the Dunkl setting in [ 41 , 42 ] (see also [ 4 , 3 , 20 , 21 ] ). On the Euclidean space \mathbb{R}^{N} we consider a normalized root system R and a multiplicity function k(\nu)\geq 0,\nu\in R . Define dw({\rm x})=\prod_{\nu\in R}|\langle{\rm x},\nu\rangle|^{k(\nu)}d{\rm x}. In this paper, for p\in(0,\infty) we denote by L^{p}(dw) the set of all measurable functions f such that \|f\|_{L^{p}(dw)}:=\Big(\int_{{\mathbb{R}^{N}}}|f({\rm x})|^{p}dw({\rm x})\Big)^{1/p}<\infty. We consider the Euclidean space {\mathbb{R}^{N}} with the usual scalar product and the induced norm. For a nonzero vector \nu\in{\mathbb{R}^{N}} the reflection \sigma_{\nu} with respect to the hyperplane \nu^{\perp} orthogonal to \nu is given by (1) \sigma_{\nu}{\rm x}={\rm x}-2\frac{\langle{\rm x},\nu\rangle}{\|\nu\|^{2}}\nu. We denote by G the Weyl group generated by the reflections \sigma_{\nu},\nu\in R . A finite set R\subset{\mathbb{R}^{N}}\backslash\{0\} is called a root system if \sigma_{\nu}(R)=R for every \nu\in R . We shall consider normalized reduced root systems, that is, \|\nu\|^{2}=2 for every \nu\in R . The finite group G generated by the reflections \sigma_{\nu} is called the Weyl group (reflection group) of the root system. A multiplicity function is a G -invariant function k:R\to[0,\infty) which will be fixed and be nonnegative throughout this paper. We will denote by \mathcal{O}({\rm x})=\{\sigma({\rm x}):\sigma\in R\} the G -orbit of a point {\rm x}\in{\mathbb{R}^{N}} , and \mathcal{O}(E)=\bigcup_{{\rm x}\in E}\mathcal{O}({\rm x}) the G -orbit of E\subset{\mathbb{R}^{N}} . The number \mathfrak{N}:=N+\sum_{\nu\in R}k(\nu) is called the homogeneous dimension of the system, since (2) w(B(t{\rm x},tr))=t^{\mathfrak{N}}w(B({\rm x},r))\ \ \text{for}\ \ {\rm x}\in{\mathbb{R}^{N}},t,r>0, where B({\rm x},r)=\{{\rm y}\in{\mathbb{R}^{N}}:\|{\rm x}-{\rm y}\|\leq r\} denotes the (closed) Euclidean ball centered at {\rm x} with radius r>0 . Observe that (3) w(B({\rm x},r))\simeq r^{N}\prod_{\nu\in R}(|\langle{\rm x},\nu\rangle|+r)^{k(\nu)}\gtrsim r^{\mathfrak{N}}, so dw is a doubling measure, that is, there exists a constant C>0 such that (4) w(B({\rm x},2r))\leq Cw(B({\rm x},r))\ \ \text{for}\ \ {\rm x}\in{\mathbb{R}^{N}},r>0. Moreover, by ( 3 ), (5) \Big(\frac{R}{r}\Big)^{N}\lesssim\frac{w(B({\rm x},R))}{w(B({\rm x},r))}\lesssim\Big(\frac{R}{r}\Big)^{\mathfrak{N}}\ \ \text{for}\ \ {\rm x}\in{\mathbb{R}^{N}},0<r<R. Finally, define (6) d({\rm x},{\rm y})=\min_{\sigma\in G}\|{\rm x}-\sigma({\rm y})\| as the distance between two G -orbits \mathcal{O}({\rm x}) and \mathcal{O}({\rm y}) . Obviously, \mathcal{O}(B({\rm x},r))=\{{\rm y}\in{\mathbb{R}^{N}}:d({\rm x},{\rm y})<r\} ; moreover, w(B({\rm x},r))\leq w(\mathcal{O}(B({\rm x},r)))\leq|G|w(B({\rm x},r)) for all {\rm x}\in{\mathbb{R}^{N}} and r>0 . We also denote B^{d}({\rm x},r):=\{{\rm y}\in{\mathbb{R}^{N}}:d({\rm x},{\rm y})<r\} , i.e., B^{d}({\rm x},r)=\mathcal{O}(B({\rm x},r)) . Let E({\rm x},{\rm y}) be the associated Dunkl kernel, which was introduced in [ 16 ] . It is known that E({\rm x},{\rm y}) has a unique extension to a holomorphic function in \mathbb{C}^{N}\times\mathbb{C}^{N} . The Dunkl transform is defined by (7) \mathcal{F}f(\xi)=c_{k}^{-1}\int_{\mathbb{R}^{N}}E(-i\xi,{\rm x})f({\rm x})dw({\rm x}), where c_{k}=\int_{\mathbb{R}^{N}}e^{-\frac{\|{\rm x}\|^{2}}{2}}dw({\rm x})>0. It is know that the Dunkl transform, which is initially defined for f\in L^{1}(dw) , is an isometry on L^{2}(dw) and preserves the Schwartz class of functions \mathcal{S}(\mathbb{R}^{N}) . The inverse of the Dunkl transform is defined by {\mathcal{F}}^{-1}g({\rm x})=c_{k}^{-1}\int_{{\mathbb{R}^{N}}}E(i\xi,{\rm x})g(\xi)dw(\xi). For {\rm x}\in{\mathbb{R}^{N}} , the Dunkl translation \tau_{{\rm x}} is defined by (8) \tau_{{\rm x}}f({\rm y})=c_{k}^{-1}\int_{{\mathbb{R}^{N}}}E(i\xi,{\rm x})E(i\xi,{\rm y}){\mathcal{F}}f(\xi)dw(\xi) is bounded on L^{2}(dw) . In what follows, we use the notation g({\rm x},{\rm y})=\tau_{{\rm x}}g(-{\rm y})=\tau_{-{\rm y}}g(x). The Dunkl convolution of two appropriate functions is defined by (9) f\ast g({\rm x})=c_{k}{\mathcal{F}}^{-1}[{\mathcal{F}}f{\mathcal{F}}g]({\rm x})=\int_{{\mathbb{R}^{N}}}{\mathcal{F}}f(\xi){\mathcal{F}}g(\xi)E({\rm x},i\xi)dw(\xi), or equivalently, f\ast g({\rm x})=\int_{{\mathbb{R}^{N}}}f({\rm y})\tau_{{\rm x}}g(-{\rm y})dw({\rm y})=\int_{{\mathbb{R}^{N}}}f({\rm y})g({\rm x},{\rm y})dw({\rm y}). We now consider the Dunkl operators T_{\xi} defined by T_{\xi}f({\rm x})=\partial_{\xi}f({\rm x})+\sum_{\nu\in R}\frac{k(\nu)}{2}\langle\nu,\xi\rangle\frac{f({\rm x})-f(\sigma_{\nu}({\rm x}))}{\langle\nu,{\rm x}\rangle}, where \partial_{\xi}f denotes the directional derivative of f in the direction \xi . Set T_{j}=T_{e_{j}} , where \{e_{1},\ldots,e_{N}\} is the canonical basis for {\mathbb{R}^{N}} . The Dunkl Laplacian is defined by \displaystyle L: \displaystyle=-\sum_{j=1}^{N}T_{j}^{2} \displaystyle=-\Delta_{{\mathbb{R}^{N}}}-\sum_{\nu\in R}k(\nu)\delta_{\nu}f({\rm x}), where \Delta_{{\mathbb{R}^{N}}} is the Laplacian on {\mathbb{R}^{N}} and \delta_{\nu} is defined by \delta_{\nu}f({\rm x})=\frac{\partial_{\nu}f({\rm x})}{\langle\alpha,{\rm x}\rangle}-\frac{f({\rm x})-f(\sigma_{\nu}({\rm x}))}{\langle\alpha,{\rm x}\rangle^{2}}. It is well-known that the operator L is a non-negative and self-adjoint operator on L^{2}(dw) and generates a semigroup e^{-tL} whose kernel h_{t}({\rm x},{\rm y}) is defined by h_{t}({\rm x},{\rm y})=c_{k}^{-1}(2t)^{-\mathfrak{N}/2}\exp\Big(-\frac{\|{\rm x}\|^{2}+\|{\rm y}\|^{2}}{4t}\Big)E\Big(\frac{{\rm x}}{\sqrt{2t}},\frac{{\rm y}}{\sqrt{2t}}\Big),\ \ \ {\rm x},{\rm y}\in{\mathbb{R}^{N}},t>0. See for example [ 2 , 38 ] . The paper focuses on two primary objectives. Firstly, we will investigate some characterizations of the Besov spaces and Triebel-Lizorkin spaces in ({\mathbb{R}^{N}},\|\cdot\|,dw) via the heat semigroup e^{-tL} . Secondly, we will prove the boundedness of the multiplier of the Dunkl transform on these function spaces. For the first aim, we would like to go back the theory of Hardy spaces in ({\mathbb{R}^{N}},\|\cdot\|,dw) . Since ({\mathbb{R}^{N}},\|\cdot\|,dw) is a homogneous type in the sense of Coifman and Weiss, one can define the Hardy spaces H^{p}_{\rm CW}(dw) with \frac{\mathfrak{N}}{\mathfrak{N}+1}<p\leq 1 via the atomic decomposition. For convenience, we provide the definition of the Hardy spaces H^{p}_{\rm CW}(dw) with \frac{\mathfrak{N}}{\mathfrak{N}+1}<p\leq 1 in [ 11 ] , which can be viewed as extensions to the Hardy spaces on \mathbb{R}^{N} in the influential paper of Fefferman and Stein [ 22 ] (see also [ 47 , 46 ] ). For 0<p\leq 1 , we say that a function a is a (p,2) atom if there exists a ball B such that (i) supp a\subset B ; (ii) \|a\|_{L^{2}(dw)}\leq w(B)^{1/2-1/p} ; (iii) \displaystyle\int a({\rm

Read the complete text