Graph Coloring Based Pilot Reuse among Interfering Users in Cell-Free Massive MIMO

Abstract

In cell-free massive multiple-input-multiple-output (CF-mMIMO) systems, we spread a massive number of controllable access points out over a coverage area to simultaneously serve much smaller number of user equipments (UEs) over the same time/frequency resources. The major objective of coining such systems is to offer to all UEs a fair quality of service, avoiding the high interference for cell-edge UEs in cellular networks. However, this is degraded by the pilot contamination problem due to the inter-user interference (IUI). In addition, contrary to the centralized mMIMO, CF-mMIMO is characterized by a channel hardening not sufficiently pronounced, thus it will be appropriate to include downlink (DL) pilots to estimate the DL channel. In this paper, we evolve a graph coloring algorithm for DL pilot assignment in CF-mMIMO system (GC-PA-CFmMIMO) to mitigate the DL pilot contamination. Especially, by exploiting IUI in DL, a conflict or interference graph aims to model the potential interference among interfering users. Then, GC-PA-CF-mMIMO mitigates the potential IUI in DL by appointing adequate pilots among UEs in the conflict graph. The simulation results validate the proposed approach and reveal that, remarkably, it outperforms the DL pilot assignment methods based on prior approaches in terms of min-per-user DL net throughput.