### abstract

- We consider a geometric optimization problem that arises in network design. Given a set P of n points in the plane, source and destination points s,t ∈ P, and an integer k > 0, one has to locate k Steiner points, such that the length of the longest edge of a bottleneck path between s and t is minimized. In this paper, we present an O(n log 2 n)-time algorithm that computes an optimal solution, for any constant k. This problem was previously studied by Hou et al. [Hou10], who gave an O(n 2 log n)-time algorithm. We also study the dual version of the problem, where a value λ > 0 is given (instead of k), and the goal is to locate as few Steiner points as possible, so that the length of the longest edge of a bottleneck path between s and t is at most λ. Our algorithms are based on two new geometric structures that we develop --- an (α,β)-pair decomposition of P and a floor (1+e)-spanner of P. For real numbers β > α > 0, an (α,β)-pair decomposition of P is a collection W={(A 1 ,B 1 ),...,(A m ,B m )} of pairs of subsets of P, satisfying: (i) For each pair (A i ,B i ) ∈ W, the radius of the minimum enclosing circle of A i (resp. B i ) is at most α, and (ii) For any p,q ∈ P, such that |pq| ≤ β, there exists a single pair (A i ,B i ) ∈ W, such that p ∈ A i and q ∈ B i , or vice versa. We construct (a compact representation of) an (α,β)-pair decomposition of P in time O((β/α) 3 n log n). Finally, for the complete graph with vertex set P and weight function w(p,q) = ⌊|pq|⌋, we construct a (1+e)-spanner of size O(n/e 4 ) in time O((1/e 4 )n log 2 n), even though w is not a metric.