欧氏空间中由带漂移的拉普拉斯算子的预解式定义的若干调和分析算子的 $L^p$-有界性性质
$L^p$-boundedness properties for some harmonic analysis operators defined by resolvents for a Laplacian with drift in Euclidean spaces
AI总结:
本文研究了欧氏空间中由带漂移拉普拉斯算子预解式定义的调和分析算子,在非双倍但局部双倍测度下的 $L^p$ 和弱 $L^1$ 有界性。
摘要:
我们考虑欧氏空间 $\mathbb R^n$ 中由 $Δ_ν= \sum_{i=1}^n(\frac{\partial^2}{\partial x_i^2} + 2 ν_i\frac{\partial }{\partial{x_i}})$ 定义的带漂移的拉普拉斯算子,其中 $ν=(ν_1,\ldots,ν_n)\in \mathbb R^n\setminus\{0\}$。算子 $Δ_ν$ 关于测度 $dμ_ν(x)=e^{2\langleν,x\rangle}dx$ 是自伴的。该测度不是双倍测度,但在 $\mathbb R^n$ 中是局部双倍的。对于每个 $M>0$ 和 $k \in \mathbb N$,我们定义算子 $$ W^k_{ν,M,*}(f) = \sup_{t>0}\left|A^k_{ν,M,t}(f)\right|,\hspace{5mm}g_{ν,M}^k(f) = \left(\int_0^\infty\left|A^k_{ν,M,t}(f)\right|^2\frac{dt}{t}\right)^{\frac{1}{2}},k\geq 1, $$ $ρ$-变差算子 $$ V_ρ\left( \{A^k_{ν,M,t}\}_{t>0}\right)(f)= \sup_{0<t_1<\cdots<t_\ell,\\,\ell \in \mathbb N}\left(\sum^{\ell-1}_{j=1}\left|A^k_{ν,M,t_j}(f)- A^k_{ν,M,t_{j+1}}(f)\right|^ρ\right)^{\frac{1}ρ},ρ>2, $$ 以及,如果 $\{t_j\}_{j\in \mathbb N}$ 是 $(0,\infty)$ 中的递减序列,则定义振荡算子 $$ O(\{A_{ν,M,t}^k\}_{t>0},\{t_j\}_{j\in \mathbb N})(f)=\Big(\sum_{j\in \mathbb N}\\;\\;\sup_{t_{j+1}\leq \varepsilon <\varepsilon '\leq t_j}|A^k_{ν,M,\varepsilon}(f)-A^k_{ν,M,\varepsilon '}(f)|^2 \Big)^{1/2}。 $$ 其中 $A^k_{ν,M,t}=t^k\partial^k_t(I-tΔ_ν)^{-M}$,$t>0$。我们用 $T_{ν,M}^k$ 表示上述任意一个算子。我们研究了 $T^k_{ν,M}$ 从 $L^p(\mathbb R^n,μ_ν)$ 到其自身的有界性(对于每个 $1<p<\infty$),以及从 $L^1(\mathbb R^n,μ_ν)$ 到 $L^{1,\infty}(\mathbb R^n,μ_ν)$ 的有界性。此外,我们还得到了算子 $G_{ν,M}^{k,\ell}$($1\leq \ell <2M$)的有界性性质,其定义为 $$ G_{ν,M}^{k,\ell}(f)=\left(\int_0^\infty\left|t^{\ell /2+k}\partial _t^kD^{(\ell)}(I-tΔ_ν)^{-M}(f) \right|^2\frac{dt}{t}\right)^{\frac{1}{2}}, $$ 其中 $D^{(\ell)}$ 为某个微分算子。
英文摘要:
We consider the Laplacian with drift in $\mathbb R^n$ defined by $Δ_ν= \sum_{i=1}^n(\frac{\partial^2}{\partial x_i^2} + 2 ν_i\frac{\partial }{\partial{x_i}})$ where $ν=(ν_1,\ldots,ν_n)\in \mathbb R^n\setminus\{0\}$. The operator $Δ_ν$ is selfadjoint with respect to the measure $dμ_ν(x)=e^{2\langleν,x\rangle}dx$. This measure is not doubling but it is locally doubling in $\mathbb R^n$. We define, for every $M>0$ and $k \in \mathbb N$, the operators $$ W^k_{ν,M,*}(f) = \sup_{t>0}\left|A^k_{ν,M,t}(f)\right|,\hspace{5mm}g_{ν,M}^k(f) = \left(\int_0^\infty\left|A^k_{ν,M,t}(f)\right|^2\frac{dt}{t}\right)^{\frac{1}{2}},\,k\geq 1, $$ the $ρ$-variation operator $$ V_ρ\left( \{A^k_{ν,M,t}\}_{t>0}\right)(f)= \sup_{0<t_1<\cdots<t_\ell,\,\ell \in \mathbb N}\left(\sum^{\ell-1}_{j=1}\left|A^k_{ν,M,t_j}(f)- A^k_{ν,M,t_{j+1}}(f)\right|^ρ\right)^{\frac{1}ρ},\;\; ρ>2, $$ and, if $\{t_j\}_{j\in \mathbb N}$ is a decreasing sequence in $(0,\infty)$, the oscillation operator $$ O(\{A_{ν,M,t}^k\}_{t>0},\{t_j\}_{j\in \mathbb N})(f)=\Big(\sum_{j\in \mathbb N}\;\;\sup_{t_{j+1}\leq \varepsilon <\varepsilon '\leq t_j}|A^k_{ν,M,\varepsilon}(f)-A^k_{ν,M,\varepsilon '}(f)|^2 \Big)^{1/2}. $$ where $A^k_{ν,M,t}=t^k\partial^k_t(I-tΔ_ν)^{-M}$, $t>0$. We denote by $T_{ν,M}^k$ any of the above operators. We analyze the boundedness of $T^k_{ν,M}$ on $L^p(\mathbb R^n,μ_ν)$ into itself, for every $1<p<\infty$, and from $L^1(\mathbb R^n,μ_ν)$ into $L^{1,\infty}(\mathbb R^n,μ_ν)$. In addition, we obtain boundedness properties for the operator $G_{ν,M}^{k,\ell}$, $1\leq \ell <2M$, defined by $$ G_{ν,M}^{k,\ell}(f)=\left(\int_0^\infty\left|t^{\ell /2+k}\partial _t^kD^{(\ell)}(I-tΔ_ν)^{-M}(f) \right|^2\frac{dt}{t}\right)^{\frac{1}{2}}, $$ for certain differentiation operator $D^{(\ell)}$.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。