Cooperative Base Station Coloring for Pair-wise Multi-Cell Coordination

TL;DR

This paper introduces a novel BS clustering and pattern assignment method using Voronoi regions and edge-coloring, which prevents conflicts and guarantees dual-BS communication, improving multi-cell coordination especially in dense networks.

Contribution

The paper presents a new pair-wise BS coordination approach based on Voronoi regions and graph edge-coloring, ensuring conflict-free BS selection in irregular topologies.

Findings

Achieves conflict-free BS selection in irregular networks.

Provides analytical expressions for rate distribution and spectral efficiency.

Performs comparably to dynamic clustering in dense user scenarios.

Abstract

This paper proposes a method for designing BS clusters and cluster patterns for pair-wise BS coordination. The key idea is that each BS cluster is formed by using the 2nd-order Voronoi region, and the BS clusters are assigned to a specific cluster pattern by using edge-coloring for a graph drawn by Delaunay triangulation. The main advantage of the proposed method is that the BS selection conflict problem is prevented, while selected users are guaranteed to communicate with their two closest BSs in any irregular BS topology. With the proposed coordination method, analytical expressions for the rate distribution and the ergodic spectral efficiency are derived as a function of relevant system parameters in a fixed irregular network model. In a random network model with a homogeneous Poisson point process, a lower bound on the ergodic spectral efficiency is characterized. Through system level simulations, the performance of the proposed method is compared with that of conventional coordination methods: dynamic clustering and static clustering. Our major finding is that, when users are dense enough in a network, the proposed method provides the same level of coordination benefit with dynamic clustering to edge users.