A geometric graph is a simple graph <i>G=(V,E)</i> with an embedding of the set <i>V</i> in the plane such that the points that represent <i>V</i> are in general position. A geometric graph is said to be <i>k-locally Delaunay</i> (or a <i>D</i><sub>k</sub>-graph) if for each edge <i>(u,v)</i> ∈ <i>E</i> there is a (Euclidean) disc <i>d</i> that contains <i>u</i> and <i>v</i> but no other vertex of <i>G</i> that is within <i>k</i> hops from <i>u</i> or <i>v</i>.The study of these graphs was recently motivated by topology control for wireless networks [6,7]. We obtain the following results: (i) We prove that if <i>G</i> is a <i>D</i><sub>1</sub>-graph on <i>n</i> vertices, then it has <i>O</i>(<i>n</i><sup>3/2</sup>) edges. (ii) We show that for any <i>n</i> there exist <i>D</i><sub>1</sub>-graphs with n vertices and Ω(n<sup>4/3</sup>) edges. (iii) We prove that if G is a D<sub>2</sub>-graph on n vertices, then it has O(n) edges. This bound is worst-case asymptotically tight. As an application of the first result, we show that: (iv) The maximum size of a family of pairwise non-overlapping lenses in an arrangement of $n$ unit circles in the plane is O(n<sup>3/2</sup>).The first two results improve the best previously known upper and lower bounds of $O(n^ 5/3 )$ and $\Omega(n)$ respectively (see \cite KL03 ). The third result improves the best previously known upper bound of O(n log n ) ([6]). Finally, our last result improves the best previously known upper bound (for the more general case of not necessarily unit circles) of O(n<sup>3/2</sup> κ(n)) (see [1] ), where κ(n) = (log n ) <sup>O(α<sup>2</sup>(n))</sup> and where α(n) is the extremely slowly growing inverse Ackermann's function.
[1]
János Pach,et al.
Geometric graphs with no self-intersecting path of length three
,
2004,
Eur. J. Comb..
[2]
Herbert Edelsbrunner,et al.
On the maximal number of edges of many faces in an arrangement
,
1986,
J. Comb. Theory, Ser. A.
[3]
Rom Pinchasi,et al.
Topological graphs with no self-intersecting cycle of length 4
,
2003,
SCG '03.
[4]
Noga Alon,et al.
On the Complexity of Arrangements of Circles in the Plane
,
2001,
Discret. Comput. Geom..
[5]
J. Pach,et al.
Geometric graphs with no self-intersecting path of length three
,
2002,
Eur. J. Comb..
[6]
Micha Sharir,et al.
Lenses in arrangements of pseudo-circles and their applications
,
2004,
JACM.
[7]
Xiang-Yang Li,et al.
Distributed construction of a planar spanner and routing for ad hoc wireless networks
,
2002,
Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.
[8]
M. Sharir,et al.
Lenses in arrangements of pseudo-circles and their applications
,
2002,
SCG '02.
[9]
J. Pach.
Finite point configurations
,
1997
.
[10]
B. Bollobás,et al.
Extremal Graph Theory
,
2013
.
[11]
Timothy M. Chan.
On Levels in Arrangements of Curves
,
2003,
Discret. Comput. Geom..
[12]
Hisao Tamaki,et al.
How to Cut Pseudoparabolas into Segments
,
1998,
Discret. Comput. Geom..