On a lattice-point problem in hyperbolic space and related questions in spectral theory (Q1102997): Difference between revisions
From MaRDI portal
ReferenceBot (talk | contribs) Changed an Item |
Set OpenAlex properties. |
||
Property / full work available at URL | |||
Property / full work available at URL: https://doi.org/10.1007/bf02386116 / rank | |||
Normal rank | |||
Property / OpenAlex ID | |||
Property / OpenAlex ID: W2005000521 / rank | |||
Normal rank |
Latest revision as of 09:56, 30 July 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | On a lattice-point problem in hyperbolic space and related questions in spectral theory |
scientific article |
Statements
On a lattice-point problem in hyperbolic space and related questions in spectral theory (English)
0 references
1988
0 references
Let \(\Gamma\) be a geometrically finite discrete group of isometries of the \((N+1)\)-dimensional hyperbolic space \(H^{N+1}\) and assume that the abscissa of convergence \(\delta =\delta (\Gamma)\) of the Poincaré series for \(\Gamma\) satisfies \(\delta (\Gamma)>N/2\). For \(x,x'\in H^{N+1}\) and \(X>0\) let \(N(X;x,x')\) denote the number of lattice points \(\gamma x'\) (\(\gamma\in \Gamma)\) which lie in a large hyperbolic sphere (measured by \(X)\) with center \(x\). Then \textit{P.~D. Lax} and \textit{R.~S. Phillips} [J. Funct. Anal. 46, 280--350 (1982; Zbl 0497.30036)] proved that there exists a unique eigenfunction \(F\) of the Laplacian on \(L^2(\Gamma \setminus H^{N+1})\) with eigenvalue \(-\delta (N-\delta)\) and \(c\), \(\eta >0\) such that \[ N(X;x,x')=cF(x)F(x')X^{\delta}+O(X^{\delta -\eta})\quad\text{as } X\to \infty. \] The main objective of the paper under review is to prove analogous results when \(\delta\le N/2\). Theorem 1: If \(\Gamma\) is convex cocompact, then there exists some \(c>0\) so that \[ N(X;x,x')\sim cF(x)F(x')X^{\delta}\quad\text{as } X\to \infty. \] (Note that this theorem yields no information about the error term.) The proof of Theorem 1 is based on the spectral theory of the Laplacian on \(L^2(\Gamma \setminus H^{N+1})\), in particular, on the fact that the resolvent kernel has essentially a simple pole at \(\delta\) and no further poles on the line \(\Re s=\delta\) (see Theorem 2). The proof of Theorem 2 uses ideas from ergodic theory.
0 references
number of lattice points
0 references
eigenfunction
0 references
Laplacian
0 references
resolvent kernel
0 references
simple pole
0 references