Modularity-Based User-Centric Clustering and Resource Allocation for Ultra Dense Networks

A novel modularity-based user-centric (MUC) clustering is conceived for resource allocation in ultra dense networks (UDNs), in order to maximize the sum-rate per orthogonal resource block. The idea of MUC clustering is to decompose the UDN into several sub-networks by exploiting the inherent group structure of user equipments. In particular, we propose a modified Louvain method for MUC clustering relying on efficient resource allocation heuristics. Our numerical results show the superiority of our MUC design.