CoV-Based Metrics for Quantifying the Regularity of Hard-Core Point Processes for Modeling Base Station Locations

摘要

Base station locations in wireless networks can be modeled via repulsive random point processes with an amount of regularity that is tunable between that of a triangular lattice and that of a homogeneous Poisson point process. However, it is currently difficult to quantify this regularity, or compare different repulsive point processes. In this letter, we examine three regularity metrics based on the coefficient of variation (CoV) of geometric properties of point processes and identify the CoV of the nearest neighbour distance as the most sensitive metric. We also compare three hard-core point processes in terms of their regularity range and the density of the generated points.