English

Random Geometric Graph Diameter in the Unit Ball

Combinatorics 2011-10-05 v2 Probability

Abstract

The unit ball random geometric graph G=Gpd(λ,n)G=G^d_p(\lambda,n) has as its vertices nn points distributed independently and uniformly in the dd-dimensional unit ball, with two vertices adjacent if and only if their lpl_p-distance is at most λ\lambda. Like its cousin the Erdos-Renyi random graph, GG has a connectivity threshold: an asymptotic value for λ\lambda in terms of nn, above which GG is connected and below which GG is disconnected (and in fact has isolated vertices in most cases). In the connected zone, we determine upper and lower bounds for the graph diameter of GG. Specifically, almost always, \diamp(B)(1o(1))/λ\diam(G)\diamp(B)(1+O((lnlnn/lnn)1/d))/λ\diam_p(\mathbf{B})(1-o(1))/\lambda \leq \diam(G) \leq \diam_p(\mathbf{B})(1+O((\ln \ln n/\ln n)^{1/d}))/\lambda, where \diamp(B)\diam_p(\mathbf{B}) is the p\ell_p-diameter of the unit ball B\mathbf{B}. We employ a combination of methods from probabilistic combinatorics and stochastic geometry.

Keywords

Cite

@article{arxiv.math/0501214,
  title  = {Random Geometric Graph Diameter in the Unit Ball},
  author = {Robert B. Ellis and Jeremy L. Martin and Catherine Yan},
  journal= {arXiv preprint arXiv:math/0501214},
  year   = {2011}
}

Comments

17 pages, 4 figures; exposition revised substantially, particularly in Sections 3 and 5