Integral operators with rough kernels in variable Lebesgue spaces (Q2226636)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Integral operators with rough kernels in variable Lebesgue spaces
scientific article

    Statements

    Integral operators with rough kernels in variable Lebesgue spaces (English)
    0 references
    0 references
    0 references
    8 February 2021
    0 references
    In this paper the authors study the boundedness of an integral operator from \(L^{p(\cdot)}\) into \(L^{q(\cdot)}\) for \(\frac{1}{p(\cdot)}=\frac{1}{q(\cdot)}+\frac{\alpha}{n}\), \(0\leq \alpha \leq n\) for certain exponent functions \(p(\cdot)\) satisfying weaker conditions than the classical log-Hölder conditions.\par We denote by \(\mathcal{P}(\mathbb{R}^n)\) the family of measurable functions \(p(\cdot): \mathbb{R}^n \rightarrow [1,\infty)\), and let \(p_{-}= {\text{ess}\;\inf}\ p(x)\) and \(p_{+}= {\text{ess}\;\sup}\ p(x)\). Given \(p \in \mathcal{P}(\mathbb{R}^n)\), we say that \(p \in N_{\infty}(\mathbb{R}^n)\) if there exist constants \(\Lambda_{\infty}\) and \(p_{\infty}\) such that \[\int_{D_{+}}\exp(-\Lambda_{\infty}\vert \frac{1}{p(x)}-\frac{1}{p_{\infty}}\vert ^{-1})dx < \infty,\] where \(D_{+}=\{x \in \mathbb{R}^n: \vert \frac{1}{p(x)}-\frac{1}{p_{\infty}}\vert > 0 \}\), and \(p \in K_{0}(\mathbb{R}^n)\) if there exists a constant \(C\) such that for every cube \(Q\) \[\vert \vert \chi_{Q}\vert \vert_{L^{p(\cdot)}(\mathbb{R}^n)}\vert \vert \chi_{Q}\vert \vert_{L^{p^\prime(\cdot)}(\mathbb{R}^n)} \leq C\vert Q\vert .\] Let \(m \in \mathbb{N}\). For \(1 \leq i \leq m\), let \(1 < q_{i} < \infty\) be such that \(\frac{n}{q_{1}}+\cdots +\frac{n}{q_{m}}=n-\alpha\), \(0 \leq \alpha < n\). For \(\alpha =0\) we take \(m > 1\). We denote by \(\Sigma\) the unit sphere in \(\mathbb{R}^n\). Let \(\Omega_{i} \in L^1(\Sigma)\), \(1 \leq i \leq m\) as \(\Omega_{i}(x)=\Omega_{i}(x^\prime)\) for \(x=\vert x\vert x^\prime \in \mathbb{R}^n\setminus \{0 \}\) and let \(K(x,y)=k_{1}(x-A_{1}y)\cdots k_{m}(x-A_{m}y)\), where \(k_{i}(x)=\Omega_{i}(x)/\vert x\vert ^{n/q_{i}}\) and \(A_{i}\) are invertible matrices such that \(A_{i}-A_{j}\) is invertible for \(i \neq j\), \(1 \leq i\), \(j \leq m\). Let \(T_{\alpha}\) be the integral operator with the kernel \(K(x,y)\), that is, \[T_{\alpha}f(x)=\int_{\mathbb{R}^n}K(x,y)f(y)\ dy.\] The authors prove the following main theorem: Assume the functions \(\Omega_{i}\), \(1 \leq i \leq m\), satisfy the following conditions: (1) there exist \(p_{i}> q_{i}\) such that \(\Omega_{i} \in L^{p_{i}}(\Sigma)\), (2) \(\int_0^1 \sup_{\vert y\vert < t}\vert \vert \Omega_{i}(\cdot+y)-\Omega_{i}(\cdot)\vert \vert_{L^{p_{i}}(\Sigma)}\frac{dt}{t} < \infty\). Let \(s \geq 1\) be defined by \(\frac{1}{p_{1}}+ \cdots \frac{1}{p_{m}}+\frac{1}{s}=1\) and let \(p \in \mathcal{P}(\mathbb{R}^n)\) be such that \(1 \leq s < p_{-} \leq p_{+} < \frac{n}{\alpha}\) and such that \(p(A_{i}x) \leq p(x)\) a.e., \(x \in \mathbb{R}^n\) and let \(q \in \mathcal{P}(\mathbb{R}^n)\) be defined by \(\frac{1}{p(\cdot)}-\frac{1}{q(\cdot)}=\frac{n}{\alpha}\). If \(\frac{q(\cdot)}{s} \in N_{\infty}(\mathbb{R}^n) \cap K_{0}(\mathbb{R}^n)\) then \(T_{\alpha}\) extends to a bounded operator from \(L^{p(\cdot)}(\mathbb{R}^n)\) into \(L^{q(\cdot)}(\mathbb{R}^n)\).
    0 references
    variable Lebesgue space
    0 references
    integral operator
    0 references
    rough kernel
    0 references
    maximal operator
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references