Extremal Bases, Geometrically Separated Domains and Applications

TL;DR

This paper introduces extremal bases and geometrically separated domains in complex analysis, providing new tools for sharp estimates of Bergman and Szeg"o projections, with applications to various domain classes.

Contribution

It defines extremal bases and geometrically separated domains, establishes their properties, and applies these concepts to obtain sharp projection estimates, extending previous results.

Findings

Sharp estimates for Bergman and Szeg"o projections are valid in geometrically separated domains.

Examples include locally lineally convex and diagonally Levi form domains.

Results improve upon previous estimates by Fefferman, Kohn, and Machedon.

Abstract

We introduce the notion of extremal basis of tangent vector fields at a boundary point of finite type of a pseudo-convex domain in $\mathbb{C}^n$. Then we define the class of geometrically separated domains at a boundary point, and give a description of their complex geometry. Examples of such domains are given, for instance, by locally lineally convex domains, domains with locally diagonalizable Levi form, and domains for which the Levi form have comparable eigenvalues at a point. Moreover we show that these domains are localizable. Then we define the notion of "adapted pluri-subharmonic function" to these domains, and we give sufficient conditions for his existence. Then we show that all the sharp estimates for the Bergman ans Szeg\"o projections are valid in this case. Finally we apply these results to the examples to get global and local sharp estimates, improving, for examlple, a result of Fefferman, Kohn and Machedon on the Szeg\"o projection.