A geometric characterisation of the quadratic min-power centre

TL;DR

This paper offers a geometric characterization of the quadratic min-power centre in the plane, introducing a linear-time algorithm for its construction and analyzing its approximation properties.

Contribution

It provides a complete geometric description of the quadratic min-power centre using Voronoi diagrams and Delaunay triangulations, along with a new efficient algorithm for its computation.

Findings

Linear-time algorithm for min-power centre construction

Geometric characterization using Voronoi diagrams and Delaunay triangulations

Performance bounds for centroid approximation

Abstract

For a given set of nodes in the plane the min-power centre is a point such that the cost of the star centred at this point and spanning all nodes is minimised. The cost of the star is defined as the sum of the costs of its nodes, where the cost of a node is an increasing function of the length of its longest incident edge. The min-power centre problem provides a model for optimally locating a cluster-head amongst a set of radio transmitters, however, the problem can also be formulated within a bicriteria location model involving the 1-centre and a generalized Fermat-Weber point, making it suitable for a variety of facility location problems. We use farthest point Voronoi diagrams and Delaunay triangulations to provide a complete geometric description of the min-power centre of a finite set of nodes in the Euclidean plane when cost is a quadratic function. This leads to a new linear-time algorithm for its construction when the convex hull of the nodes is given. We also provide an upper bound for the performance of the centroid as an approximation to the quadratic min-power centre. Finally, we briefly describe the relationship between solutions under quadratic cost and solutions under more general cost functions.