Maximal operator and its commutators on generalized weighted Orlicz-Morrey spaces (Q2414061)

In this paper, one investigates the boundedness of the maximal operator $M$ and its commutators $M^{b}$ on generalized weighted Orlicz-Morrey spaces. The results obtained provide conditions for the boundedness not in integral terms but in less restrictive terms of supremal operators. \par To state the main theorems of the paper some definitions are required. \par A function $\Phi: [0,\infty)\to [0,\infty]$ is called a Young function, if $\Phi$ is convex, left-continuous, $\lim\limits_{r\to 0^{+}}\Phi(r)=\Phi(0)=0$ and $\lim\limits_{r\to\infty}\Phi(r)=\infty$. \par The space $A_{p}$ of Muckenhoupt's weights appears for $1\le p\le\infty$ and specially for the index $p=i_{\phi}$ where \[ i_{\Phi}=\lim_{t\to 0^{+}}\frac{\log h_{\Phi}(t)}{\log t}\quad \text{and}\quad h_{\Phi}(t)=\sup_{s>0}\frac{\Phi(st)}{\Phi(s)}. \] A Young function $\Phi$ is said to satisfy the $\Delta_{2}$-condition, denoted by $\Phi\in \Delta_{2}$, if \[ \Phi (2r)\le k\Phi (r),\quad r>0 \] for some $k>1$. A Young function $\Phi$ is said to satisfy the $\nabla_{2}$-condition, denoted also by $\Phi\in\nabla_{2}$, if \[ \Phi(r)\le \frac{1}{2k}\Phi (kr),\quad r\ge 0 \] for some $k>1$. \par For a Young function $\Phi$ and a weight $w\in A_{\infty}$ set \[ L_{w}^{\Phi}(\mathbb{R}^{n})\equiv \left\{ f\text{-measurable}:\int_{\mathbb{R}_{n}}\Phi(k|f(x)|)w(x)\,dx<\infty\text{ for some }k>0\right\}. \] With these preliminaries one can define the generalized weighted Orlicz-Morrey spaces: \par Let $\varphi$ be a positive measurable function on $\mathbb{R}^{n}\times (0,\infty)$, let $w$ be a non-negative measurable function on $\mathbb{R}^{n}$ and $\Phi$ any Young function. Denote by $M_{w}^{\Phi,\varphi}(\mathbb{R}^{n})$ the generalized weighted Orlicz-Morrey space, the space of all functions $f\in L_{w}^{\Phi,\text{loc}}(\mathbb{R}^{n})$ such that \[ \begin{split} \|f\|_{M_{w}^{\Phi,\varphi}(\mathbb{R}^{n})}\equiv \|f\|_{M_{w}^{\Phi,\varphi}}&= \sup_{x\in\mathbb{R}^{n},\,r>0} \varphi(x,r)^{-1}\Phi^{-1}(w(B(x,r))^{-1})\|f\|_{L_{w}^{\Phi}(B(x,r))}\\ &= \sup_{B\in\mathcal{B}} \varphi(B)^{-1}\Phi^{-1}(w(B)^{-1})\|f\|_{L_{w}^{\Phi}(B)}<\infty. \end{split} \] Also, for a Young function $\Phi$ and a non-negative measurable function $w$, we denote by $\mathcal{G}_{\Phi}^{w}$ the set of all almost decreasing functions $\varphi : \mathbb{R}^{n}\times (0,\infty)\to (0,\infty)$ such that \[ \inf_{B\in\mathcal{B};\, r_{B}\le r_{B_{0}}}\varphi(B) \gtrsim \varphi (B_{0})\quad \text{for all }B_{0}\in\mathcal{B} \] and \[ \inf_{B\in\mathcal{B};\, r_{B}\ge r_{B_{0}}}\frac{\varphi(B)}{\Phi^{-1}(w(B)^{-1})}\gtrsim \frac{\varphi(B_{0})}{(w(B_{0})^{-1})}\quad \text{for all }B_{0}\in\mathcal{B}, \] where $r_{B}$ and $r_{B_{0}}$ denote the radius of the balls $B$ and $B_{0}$, respectively. \par The Hardy-Littlewood maximal operator $M$ is defined by \[ Mf(x)=\sup_{r>0}\frac{1}{|B(x,r)|} \int_{B(x,r)}|f(y)|\,dy,\quad x\in\mathbb{R}^{n} \] for a locally integrable function $f$ on $\mathbb{R}^{n}$. \par The following theorem for the boundedness of $M$ is proved: \par $\Phi$ be a Young function and $\varphi_{1}$, $\varphi_{2}$ positive measurable functions on $\mathbb{R}^{n}\times (0,\infty)$. \begin{itemize} \item If $\Phi\in\nabla_{2}$ and $w\in A_{i_{\Phi}}$, then the condition \[ \sup_{r<t<\infty}\left(\text{ess inf}_{t<s<\infty} \frac{\varphi_{1}(x,s)}{\Phi^{-1(w(B(x,s))^{-1})}}\right)\Phi^{-1} (w(B(x,t))^{-1})\le C\varphi_{2}(x,r), \] where $C$ does not depend on $x$ and $r$, is sufficient for the boundedness of $M$ from $M_{w} ^{\Phi,\varphi_{1}}(\mathbb{R}^{n})$ to $M^{\Phi,\varphi_{2}}_{w}(\mathbb{R}^{n})$. \par \item If $\varphi_{1}\in\mathcal{G}_{w}^{\Phi}$, then the condition \[ \varphi_{1}(x,r)\le C\varphi_{2}(x,r), \tag{1} \] where $C$ does not depend on $x$ and $r$, is necessary for the boundedness of $M$ from $M_{w}^{\Phi,\varphi_{1}}(\mathbb{R}^{n})$ to $M^{\Phi,\varphi_{2}}_{w}(\mathbb{R}^{n})$. \par \item Let $\Phi\in\nabla_{2}$ and $w\in A_{i_{\Phi}}$. If $\varphi_{1}\in\mathcal{G}_{w}^{\Phi}$, then the condition (1) is necessary and sufficient for the boundedness of $M$ from $M_{w}^{\Phi,\varphi_{1}}(\mathbb{R}^{n})$ to $M^{\Phi,\varphi_{2}}_{w}(\mathbb{R}^{n})$. \end{itemize} \par Given a measurable function $b$ the maximal commutator is defined by \[ M^{b} f(x)=\sup_{t>0}\frac{1}{|B(x,t)|} \int_{B(x,t)}|b(x)-b(y)| |f(y)|\,dy,\quad x\in\mathbb{R}^{n} \] for a locally integrable function $f$ on $\mathbb{R}^{n}$. \par Necessary and sufficient conditions for the boundedness of $M^{b}$ in generalized weighted Orlicz-Morrey spaces are given in the following result. \par Let $b\in BMO(\mathbb{R}^{n})$, $\Phi$ be a Young function and $\varphi_{1}$, $\varphi_{2}$ positive measurable functions on $\mathbb{R}^{n}\times (0,\infty)$. \par \begin{itemize} \item Let $\Phi\in \Delta_{2}\cap\nabla_{2}$ and $w\in A_{1}$, then the condition \[ \sup_{r<t<\infty}\left(1+\ln \frac{t}{r}\right) \Phi^{-1} (w(B(x,t))^{-1})\text{ess inf}_{t<s<\infty} \frac{\varphi_{1}(x,s)}{\Phi^{-1(w(B(x,s))^{-1})}}\le C\varphi_{2}(x,r), \] where $C$ does not depend on $x$ and $r$, is sufficient for the boundedness of $M^{b}$ from $M_{w}^{\Phi,\varphi_{1}}(\mathbb{R}^{n})$ to $M^{\Phi,\varphi_{2}}_{w}(\mathbb{R}^{n})$. \par \item If $\Phi\in \Delta_{2}$, $\varphi_{1}\in\mathcal{G}_{w}^{\Phi}$ and $w\in A_{1}$, then the condition (1) is necessary for the boundedness of $M^{b}$ from $M_{w}^{\Phi,\varphi_{1}}(\mathbb{R}^{n})$ to $M^{\Phi,\varphi_{2}}_{w}(\mathbb{R}^{n})$. \par \item Let $\Phi\in \Delta_{2}\cap \nabla_{2}$ and $w\in A_{1}$. If $\varphi_{1}\in\mathcal{G}_{w}^{\Phi}$ satisfies the condition \[ \sup_{r<t<\infty}\left(1+\ln\frac{t}{r}\right)\varphi_{1}(x,t)\le C\varphi_{1}(x,r), \] where $C$ does not depend on $x$ and $r$, then the condition (1) is necessary and sufficient for the boundedness of $M^{b}$ from $M_{w}^{\Phi,\varphi_{1}}(\mathbb{R}^{n})$ to $M^{\Phi,\varphi_{2}}_{w}(\mathbb{R}^{n})$. \end{itemize} \par In the paper, ones gives also necessary and sufficient conditions for the weak-type boundedness of the operator $M$ on generalized weighted Orlicz-Morrey spaces. One generalizes also for this class of spaces the classical vector-valued maximal inequalities due to Fefferman and Stein.