SearcharxivSearch

arXiv subjects

Remi Yvant Temgoua

Publications and source records attributed to Remi Yvant Temgoua.

11 recordsLinked to original sources

Fractional Hardy inequalities on $C^{1,1}$ open sets

Let $Ω$ be a bounded open set of class $C^{1,1}$ in $\mathbb{R}^N$ and $s\in(\frac{1}{2}, 1)$. We study a family of fractional Hardy-type inequalities \begin{equation} \frac{c_{N,s}}{2}\displaystyle\iint_{Ω\timesΩ}\frac{(u(x)-u(y))^2}{|x-y|^{N+2s}}\ dxdy-\displaystyleλ\int_Ωu^2\ dx\geq C\displaystyle\int_Ω\frac{u^2}{δ^{2s}}\ dx,~~~\quad\forallλ\in\mathbb{R},~~~~~~~(0.1) \end{equation} with $u\in C_c^\infty(Ω)$ and $C=C(Ω,s,N,λ)>0$. We show that the best constant in $(0.1)$ is achieved if and only if $λ>λ^*(s,Ω)$, for some $λ^*(s,Ω)\in\mathbb{R}$. As a by-product, we derive in particular that the best constant in Hardy inequality $μ_{N,s}(Ω)$ is achieved if and only if $μ_{N,s}(Ω)<\mathfrak{h}_{N,s}$, with $\mathfrak{h}_{N,s}$ being the best constant for the fractional Hardy inequality in the half space. Moreover, if $Ω$ is a convex open set, we obtain a lower bound for $λ^*(s,Ω)$ in terms of the volume of $Ω$. Specifically, we prove that $λ^*(s,Ω)\geq a(N,s)|Ω|^{-\frac{2s}{N}}$ with an explicit constant $a(N,s)>0$. Finally, for bounded $C^{1,1}$ domains, we prove that, for $s$ sufficiently close to $\frac{1}{2}$, the optimal Hardy constant is independent of both the geometry and the topology of $Ω$. More precisely, we establish that $μ_{N,s}(Ω)=\mathfrak{h}_{N,s}$. This behavior is in sharp contrast with the local case, where the topology/geometry of the domain strongly influences the value of the optimal constant, and reveals a new rigidity phenomenon in the nonlocal setting.

math.AP

A note on Pólya-Szegö inequality for fractional Orlicz-Sobolev seminorm in domains

In this paper, we study the effect of symmetric radial decreasing rearrangement on fractional Orlicz-Sobolev seminorm in domains. Roughly speaking, we prove that symmetric radial decreasing rearrangement can increase the fractional Orlicz-Sobolev seminorm in domains. Our result extends that of Li-Wang [Commun. Contemp. Math. 21.07 (2019): 1850059.] to the setting of fractional seminorm in domains admitting behaviors more general than powers.

math.AP

A supercritical nonlocal Neumann problem involving non-homogeneous fractional Laplacian

In this paper, we study the existence of positive non-decreasing radial solutions of a nonlocal non-standard growth problem ruled by the fractional $g$-Laplace operator with exterior Neumann condition. Our argument exploits some properties of fractional Orlicz-Sobolev spaces combined with a variational principle for nonsmooth functionals, which allows to deal with problems lacking compactness.

math.AP

Radial positive solutions for mixed local and nonlocal supercritical Neumann problem

In this paper, we establish the existence of positive non-decreasing radial solutions for a nonlinear mixed local and nonlocal Neumann problem in the ball. No growth assumption on the nonlinearity is required. We also provide a criterion for the existence of non-constant solutions provided the problem possesses a trivial constant solution.

math.AP

Symmetry breaking and multiplicity for supercritical elliptic Hamiltonian systems in exterior domains

We consider positive solutions of the following elliptic Hamiltonian systems \begin{equation} \left\{ \begin{aligned} -Δu+u&=a(x)v^{p-1}~~~\text{in}~~A_R\\ -Δv+v&=b(x)u^{q-1}~~~\text{in}~~A_R~~~~~~~~~~~~~~~~~(0.1)\\ u, v&>0~~~~~~~~~~~~~~~\text{in}~~A_R\\ u=v&=0~~~~~~~~~~~~~~~\text{on}~~\partial A_R, \end{aligned} \right. \end{equation} where $A_R=\{x\in\mathbb{R}^{N}: |x|>R\}$, $R>0$, $N>3$, and $a(x)$ and $b(x)$ are positive continuous functions. Under certain symmetry and monotonicity properties on $a(x)$ and $b(x)$, we prove that (0.1) has a positive solution for $(p,q)$ above the standard critical hyperbola, that is, $\frac{1}{p}+\frac{1}{q}<1-\frac{2}{N}$, enjoying the same symmetry and monotonicity properties as the weights $a$ and $b$. In the case when $a(x)=b(x)=1$, our result ensures multiplicity as it provides $\Big\lfloor \frac{N}{2}\Big\rfloor-1$ (being $\lfloor \frac{N}{2}\rfloor$ the floor of $\frac{N}{2}$) non-radial positive solutions provided that \begin{equation} (p-1)(q-1)>\Big(1+\frac{2N}{Λ_H}\Big)^{2}\Big(\frac{q}{p}\Big), \end{equation} where $Λ_H$ is the optimal constant in Hardy inequality for the domain $A_R$.

math.AP

A mixed local and nonlocal supercritical Dirichlet problems

In this work, we consider a mixed local and nonlocal Dirichlet problem with supercritical nonlinearity. We first establish a multiplicity result for the problem \begin{equation} Lu=|u|^{p-2}u+μ|u|^{q-2}u~~\text{in}~~Ω,~~~~~ u=0~~\text{in}~~\mathbb{R}^N\setminusΩ,~~~ (0.1) \end{equation} where $L=-Δ+(-Δ)^s$ for $s\in(0,1)$ and $Ω\subset\mathbb{R}^N$ is a bounded domain. Precisely, we show that problem (0.1) for $1<q<2<p$ has a positive solution as well as a sequence of sign-changing solutions with a negative energy for small values of $μ$. Here $u$ can be either a scalar function, or a vector valued function so that (0.1) turns into a system with supercritical nonlinearity. Moreover, whenever the domain is symmetric, we also prove the existence of symmetric solutions enjoying the same symmetry properties. We shall also prove an existence result for the supercritical Hamiltonian system \begin{equation} Lu=|v|^{p-2}v,~~~~~~~ Lv=|u|^{d-2}u+μ|u|^{q-2}u \end{equation} with the Dirichlet boundary condition on $Ω$ where $1<q<2<p, d$. Our method is variational, and in both problems the lack of compactness for the supercritical problem is recovered by working on a closed convex subset of an appropriate function space.

math.AP

A Hopf lemma for the regional fractional Laplacian

We provide a Hopf boundary lemma for the regional fractional Laplacian $(-Δ)^s_Ω$, with $Ω\subset\mathbb{R}^N$ a bounded open set. More precisely, given $u$ a pointwise or weak super-solution of the equation $(-Δ)^s_Ω u = c(x)u$ in $Ω$, we show that the ratio $u(x)/(\mathrm{dist}(x,\partialΩ))^{2s-1}$ is strictly positive as $x$ approaches the boundary $\partialΩ$ of $Ω$. We also prove a strong maximum principle for distributional super-solutions.

math.AP

On the s-derivative of weak solutions of the Poisson problem for the regional fractional Laplacian

In this paper, we analyze the $s$-dependence of the solution $u_s$ to the fractional Poisson equation $(-Δ)^s_Ω = f$ in an open bounded set $Ω\subset \mathbb{R}^N$. Precisely, we show that the solution map $(0,1)\to L^2(Ω)$, $s\mapsto u_s$ is continuously differentiable. Moreover, when $f = λ_s u_s$, we also analyze the one-sided differentiability of the first nontrivial eigenvalue of $(-Δ)^s_Ω$ regarded as a function of $s \in (0,1)$.

math.AP

Existence results for nonlocal problems governed by the regional fractional Laplacian

The aim of the present paper is to study existence results of minimizers of the critical fractional Sobolev constant on bounded domains. Under some values of the fractional parameter we show that the best constant is achieved. If moreover the underlying domain is a ball, we obtain positive radial minimizers for all possible values of the fractional parameter in higher dimension, while we impose a positive mass condition in low dimension.

math.AP

Morse index versus radial symmetry for fractional Dirichlet problems

In this work, we provide an estimate of the Morse index of radially symmetric sign changing bounded weak solutions $u$ to the semilinear fractional Dirichlet problem $$ (-Δ)^su = f(u)\qquad \text{ in $\mathcal{B}$},\qquad \qquad u = 0\qquad \text{in $\quad\mathbb{R}^{N}\setminus \mathcal{B}$,} $$ where $s\in(0,1)$, $\mathcal{B}\subset \mathbb{R}^N$ is the unit ball centred at zero and the nonlinearity $f$ is of class $C^1$. We prove that for $s\in(1/2,1)$ any radially symmetric sign changing solution of the above problem has a Morse index greater than or equal to $N+1$. If $s\in (0,1/2],$ the same conclusion holds under additional assumption on $f$. In particular, our results apply to the Dirichlet eigenvalue problem for the operator $(-Δ)^s$ in $\mathcal{B}$ for all $s\in (0,1)$, and it implies that eigenfunctions corresponding to the second Dirichlet eigenvalue in $\mathcal{B}$ are antisymmetric. This resolves a conjecture of Bañuelos and Kulczycki.

math.AP