Rate-Based Cell Range Expansion for Downlink Massive MIMO Heterogeneous Networks

This letter presents a rate-based user association using cell range expansion for downlink heterogeneous networks, where macro base stations (BSs) use massive multiple-input-multiple-output systems, and they are overlaid with dense small cells. Using stochastic geometry, we propose closed-form bias factors to improve the long-term average rate of a typical user. We show that such factors depend on the multiplexing gain and the BS average load. Simulation results show that the proposed bias factors efficiently distribute users across BSs, with significant rate coverage improvements in comparison with max-power and other existing biased user associations.