Averages Over Convex Hypersurfaces (Q4277965)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 490639
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Averages Over Convex Hypersurfaces |
scientific article; zbMATH DE number 490639 |
Statements
Averages Over Convex Hypersurfaces (English)
0 references
6 February 1996
0 references
Fourier transforms
0 references
maximal functions
0 references
\(L^ p\) boundedness
0 references
convex hypersurface
0 references
surface measure
0 references
maximal operator
0 references
0 references
0.93216455
0 references
0 references
0.9279533
0 references
0.9268342
0 references
0 references
0.91603607
0 references
0 references
0.9129863
0 references
Let \(\Sigma\) be a smooth convex hypersurface in \(\mathbb{R}^n\), \(n \geq 3\), of finite type, i.e., each tangent line makes finite order of contact with \(\Sigma\). Let \({\mathcal A}_t\) denote the averaging operator given by the formula \[ {\mathcal A}_t f(x) = \int_\Sigma f(x - ty) d \mu (y), \] where \(\mu\) is a smooth measure (relative to surface measure \(\sigma)\) with compact support \(\Sigma_0\) in \(\Sigma\). Let \({\mathcal M}\) denote the associated maximal operator, i.e., \({\mathcal M} f = \sup_{t > 0} |{\mathcal A}_tf |\). This paper is concerned with the \(L^p\) boundedness of \({\mathcal M}\). For \(\delta\) in \(\mathbb{R}^+\), let \(B(x, \delta)\) be the ``ball'' \(\{y \in \Sigma : \text{dist} (y, T_x (\Sigma)) < \delta \}\) in \(\Sigma\), where dist denotes the Euclidean distance and \(T_x (\Sigma)\) denotes the usual tangent space, considered as a subset of \(\mathbb{R}^n\). Two main theorems are proved, starting with the assumption that, for some positive \(\varepsilon, \gamma\), and \(C\), such that \(\gamma > 1/2\), \[ \Bigl [\int_{\Sigma_0} \bigl |B (x, \delta) \bigr |^\varepsilon d \sigma (x) \Bigr]^{1/ \varepsilon} \leq C \delta^\gamma, \] uniformly for \(\delta\) in \(\mathbb{R}^+\). It is shown that \({\mathcal M}\) is \(L^p\)-bounded in the following cases: first, if \(\varepsilon = 2/(2 \gamma-1)\) and \(p > 1 + (2 \gamma)^{- 1}\), and second, if \(p > 2 (1 + \varepsilon^{- 1})\). The proofs of these results require a virtuoso use of real analysis, Littlewood-Paley theory, and Caldéron-Zygmund theory.
0 references