English

A candidate to the densest packing with equal balls in the Thurston geometries

Metric Geometry 2012-10-09 v1

Abstract

The ball (or sphere) packing problem with equal balls, without any symmetry assumption, in a 33-dimensional space of constant curvature was settled by B\"or\"oczky and Florian for the hyperbolic space \HYP\HYP in \cite{BF64} and by proving the famous Kepler conjecture by Hales \cite{H} for the Euclidean space \EUC\EUC. The goal of this paper is to extend the problem of finding the densest geodesic ball (or sphere) packing for the other 33-dimensional homogeneous geometries (Thurston geometries) \SXR, \HXR, \SLR, \NIL, \SOL, \SXR,~\HXR,~\SLR,~\NIL,~\SOL, where a transitive symmetry group of the ball packing is assumed, one of the discrete isometry groups of the considered space. Moreover, we describe a candidate of the densest geodesic ball packing. The greatest density until now is 0.85327613\approx 0.85327613 that is not realized by packing with equal balls of the hyperbolic space \HYP\HYP. However, it attains e.g. at horoball packing of \bH3\overline{\bH}^3 where the ideal centres of horoballs lie on the absolute figure of \bH3\overline{\bH}^3 inducing the regular ideal simplex tiling (3,3,6)(3,3,6) by its Coxeter-Schl\"afli symbol. In this work we present a geodesic ball packing in the \SXR\SXR geometry whose density is 0.87499429\approx 0.87499429. The extremal configuration is described in Theorem 2.8, Our conjecture and further remarks are summarized in Section 3.

Keywords

Cite

@article{arxiv.1210.2202,
  title  = {A candidate to the densest packing with equal balls in the Thurston geometries},
  author = {Jen{\H}o Szirmai},
  journal= {arXiv preprint arXiv:1210.2202},
  year   = {2012}
}

Comments

19 pages 7 figures. arXiv admin note: substantial text overlap with arXiv:1206.0566