Embedding Results and Fractional p ( x , . ) \displaystyle p(x,.) -Laplacian Problem in Unbounded Domains

A. BARBARA 1 , A. BOUSMAHA 1 , and M. SHIMI 2 · 2025

Read Embedding Results and Fractional p ( x , . ) \displaystyle p(x,.) -Laplacian Problem in Unbounded Domains by A. BARBARA 1 , A. BOUSMAHA 1 , and M. SHIMI 2 (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-2509-01783.txt.

Introduction

Abdelkrim Barbara, Ahmed Bousmaha, and Mohammed Shimi prove a new embedding result for fractional function spaces on unbounded domains. Embedding here means one function space sits inside another with control on size. Their spaces use variable exponents, where the power in the size measure changes from point to point. Unbounded domains stretch to infinity, which removes compactness tools that many proofs rely on.

As a use case, they study nonlocal elliptic problems driven by a fractional operator of the same variable type. Nonlocal means the value at a point depends on values far away. With variational methods, meaning methods that turn equations into energy problems, they prove nontrivial weak solutions exist. Growth and smoothness conditions on the nonlinear term complete the result.

Catalog record

CREATORA. BARBARA 1 , A. BOUSMAHA 1 , and M. SHIMI 2
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE60 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-01783

Browse the full library · About the Tower of Babel

Full text

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

Embedding Results and Fractional p ( x , . ) \displaystyle p(x,.) -Laplacian Problem in Unbounded Domains A. BARBARA 1 , A. BOUSMAHA 1 , and M. SHIMI 2 Abstract . In this paper, we prove a new continuous embedding theorem for fractional Sobolev spaces with variable exponents into variable exponent Lebesgue spaces on unbounded domains. As an application, we study a class of nonlocal elliptic problems driven by the fractional p ⁡ ( x , ⋅ ) \displaystyle p(x,\cdot) -Laplacian operator. Using variational methods combined with the established embedding result, we prove the existence of nontrivial weak solutions under suitable growth and regularity conditions on the nonlinearity. A significant analytical challenge addressed in this work arises from both the nonlocal behavior of the fractional p ⁡ ( x , ⋅ ) \displaystyle p(x,\cdot) -Laplacian and the lack of compactness induced by the unboundedness of the domain. 2010 Mathematics Subject Classification 35R11, 47G20, 35S15, 35A15. 1. Introduction and statement of the main results Lebesgue and Sobolev spaces ( \displaystyle L^{p},W^{m,p}\text{ with }m\in\mathbb{N}, [ 1 ] ) were and are the backbone (in the widest sense of the word) of the theory of function spaces and PDE’s from the very beginning up to our time. However, classical Lebesgue and Sobolev spaces face limitations when dealing with nonlinear problems in applied sciences. To overcome these challenges, variable exponent Lebesgue-Sobolev spaces \displaystyle(L^{p(x)},W^{m,p(x)}) were introduced. These spaces are not only important in theoretical studies but also useful for modeling problems with variable exponent growth, which arise in various scientific applications (see for instance [ 9 , 13 , 14 ] ). Another significant extension is provided by fractional Sobolev spaces \displaystyle W^{s,p} , with \displaystyle s\in(0,1) [ 10 ] , play a key role in functional and harmonic analysis and have been widely applied in real-world problems. As a continuation of these advancements, many scholars combined the variable exponent setting with fractional case for introducing the fractional Sobolev spaces with variable exponent \displaystyle W^{s,p(x,y)} which opening new perspectives in functional and harmonic analysis and its applications. The qualitative functional properties, as well as some continuous and compact embedding theorems of these spaces into variable exponent Lebesgue spaces and Hölder spaces, have been established. Precisely, Kaufmann et al. [ 12 ] introduced the new fractional Sobolev spaces with variable exponents \displaystyle W^{s,q(x),p(x,y)}(\Omega) with the restriction \displaystyle p(x,x)<q(x) , that takes into account a fractional variable exponent operator \displaystyle(-\Delta)_{p(x,.)}^{s} . Moreover, they established a continuous and compact embedding theorem of these spaces into variable exponent Lebesgue spaces. As an application, they studied the existence and uniqueness of a solution for a nonlocal problem involving the fractional \displaystyle p(x,.) -Laplacian operator. In this context, the authors in [ 7 ] investigated fundamental properties of these function spaces and the related nonlocal operator \displaystyle(-\Delta)_{p(x,.)}^{s} . They also show the existence of at least one solution for equations involving the fractional \displaystyle p(x,.) -Laplacian. Subsequently, Ho and Kim [ 11 ] proved some fundamental embedding results for the fractional Sobolev space with variable exponent in the case \displaystyle p(x,x)\leq q(x) along with applications such as a priori bounds and multiplicity of solutions of the fractional \displaystyle p(x,\cdot) -Laplacian problems. The authors in [ 8 ] considered the case \displaystyle q(x)=p(x,x) and established a trace theorem in fractional spaces with variable exponents. Accordingly, In [ 2 ] , the authors considered the case \displaystyle p(x,x)=q(x) which represents the true generalization of the fractional Sobolev space \displaystyle W^{s,p} in the constant case. Afterward, they provided a new version of continuous and compact embedding of theses spaces into generalized Lebesgue spaces. In [ 5 ] they further established an embedding results into Hölder spaces \displaystyle\mathscr{C}^{0,r}_{b} . Since The space \displaystyle W^{s,p(x,y)}(\Omega) is not suitable for studying the fractional \displaystyle p(x,.) -Laplacian problems with Dirichlet boundary data \displaystyle u=0 in \displaystyle\mathbb{R}^{N}\setminus\Omega via variational methods and hence, [ 3 , 6 ] introduced a more appropriate space, denoted by \displaystyle W^{s,p(x,y)}(Q) . For this new space \displaystyle W^{s,p(x,y)}(Q) , they established their qualitative properties and compared it with the space \displaystyle W^{s,p(x,y)}(\Omega) Furthermore, they extended all the continuous and compact embedding results previously established for \displaystyle W^{s,p(x,y)}(\Omega) into the spaces \displaystyle W^{s,p(x,y)}(Q) and \displaystyle W_{K}^{s,p(x,y)}(Q) . To the best of our knowledge, all the aforementioned results have been considered in the case of a bounded open domain \displaystyle\Omega of \displaystyle\mathbb{R}^{N} . This naturally raises the question : what happens if we replace the bounded domain with an unbounded one ? As in the classical setting, studying nonlocal problems in unbounded domains presents several challenges compared to bounded domains. One of the main difficulties is the lack of compactness, as many fundamental functional analysis tools, such as the Rellich-Kondrachov compactness theorem, no longer hold. As a result, weak convergence results become more difficult to establish in this framework. Hence, our main aim in this paper is to start addressing these new challenges by considering an unbounded domain \displaystyle\Omega with cone property . Our goal is to establish an embedding theorem and discuss the existence of nontrivial solutions for a fractional \displaystyle p(x,\cdot) -Laplacian problem in this setting. Now, Given an unbouded domain \displaystyle\Omega with cone condition (see Definition 2.1 ). We start by fixing \displaystyle s\in(0,1) and let \displaystyle p:\overline{\Omega}\times\overline{\Omega}\longrightarrow(1,+\infty) be an uniformly continuous function such that (1.1) 1<p^{-}=\inf_{(x,y)\in\overline{\Omega}\times\overline{\Omega}}p(x,y)\leq p(x,y)\leq p^{+}=\sup_{(x,y)\in\overline{\Omega}\times\overline{\Omega}}p(x,y)<+\infty and (1.2) p\text{ is symmetric, that is, }p(x,y)=p(y,x)\quad\text{for all }(x,y)\in\overline{\Omega}\times\overline{\Omega}. We set \overline{p}(x)=p(x,x)\quad\text{for all }x\in\overline{\Omega}. In order to state our main embedding result, we define the fractional Sobolev space with variable exponent via the Gagliardo approach as follows : \displaystyle\displaystyle W \displaystyle\displaystyle=W^{s,p(x,y)}(\Omega) \displaystyle\displaystyle=\left\{u\in L^{\overline{p}(x)}(\Omega)\text{ :}\,\int_{\Omega\times\Omega}\frac{|u(x)-u(y)|^{p(x,y)}}{\lambda^{p(x,y)}|x-y|^{N+sp(x,y)}}\,dxdy<\infty,\,\,\text{for some }\lambda>0\right\}, where \displaystyle L^{\overline{p}(x)}(\Omega) is the Lebegue space with variable exponent (see section 2 ). The space \displaystyle W^{s,p(x,y)}(\Omega) is a Banach space (see [ 2 , 8 , 12 ] ) if it is equipped with the norm \|u\|_{W}=\|u\|_{L^{\overline{p}(x)}(\Omega)}+[u]_{s,p(x,y)}, where \displaystyle[.]_{s,p(x,y)} is a Gagliardo seminorm with variable exponent, which is defined by [u]_{s,p(x,y)}=[u]_{s,p(x,y)}(\Omega)=\inf\left\{\lambda>0:\int_{\Omega}\int_{\Omega}\frac{|u(x)-u(y)|^{p(x,y)}}{\lambda^{p(x,y)}|x-y|^{N+sp(x,y)}}\,dxdy<1\right\}. The space \displaystyle(W,\|.\|_{W}) is a separable reflexive [ 7 , Lemma 3.1] . In classical Sobolev space theory, embedding results play a fundamental role, particularly in functional analysis and the study of fractional Sobolev spaces. These results are essential for establishing the existence, uniqueness, and regularity of solutions to nonlocal problems, providing a rigorous framework for analyzing the qualitative behavior of solutions and ensuring their well-posedness. In this context, we present our first main result : an embedding theorem that serves as a key tool for addressing such nonlocal problems. Theorem 1.1 . Let \displaystyle\Omega\subset\mathbb{R}^{N} be an unbounded domain with cone condition and \displaystyle s\in(0,1) . Let \displaystyle p:\overline{\Omega}\times\overline{\Omega}\longrightarrow(1,+\infty) be an uniformly continuous function satisfying \displaystyle(\ref{n}) and \displaystyle(\ref{l}) with \displaystyle sp^{+}<N . Assume that \displaystyle q:\overline{\Omega}\longrightarrow(1,+\infty) is an uniformly continuous function such that (1.3) \overline{p}(x)\leq q(x)\qquad\text{ for all }x\in\overline{\Omega} and (1.4) \inf_{x\in\overline{\Omega}}\left(p_{s}^{*}(x)-q(x)\right)>0. Then \displaystyle W^{s,p(x,y)}(\Omega) is continuously embedded in \displaystyle L^{q(x)}(\Omega) . Based on this embedding result, we aim to investigate the following nonlocal problem associated with the fractional \displaystyle p(x,.) -Laplacian : (\mathcal{P}_{s})~~\left\{\begin{array}[]{rll}\left(-\Delta\right)_{p(x,.)}^{s}u(x)+b(x)|u(x)|^{\overline{p}(x)-2}\,u(x)&=f(x,u)&\text{in }\Omega,\\ u&=0&\text{in }\mathbb{R}^{N}\backslash\Omega,\end{array}\right. where : • \displaystyle\Omega is an unbounded domain in \displaystyle\mathbb{R}^{N} with cone property. • \displaystyle 0<b_{0}\leq b(x)\in L^{\infty}(\Omega) . • \displaystyle p:\overline{\Omega}\times\overline{\Omega}\longrightarrow(1,+\infty) is an uniformly continuous function satisfying ( 1.1 )-( 1.4 ). • The fractional \displaystyle

Read the complete text