A Novel PHP-Based Coverage Analysis in Millimeter Wave Heterogeneous Cellular Networks

Abstract

Due to the sensitivity of millimeter wave (mmWave) signals to blockage and using directional beamforming in both transmitter and receiver, novel models are needed to analyze cellular networks operating in mmWave frequencies. Base station (BS) locations in heterogeneous cellular networks (HCNs) is often modeled by a Poisson point process (PPP). While it is a proper model due to the randomness of BS locations, it cannot accurately model the network when we have multi-tiered BSs, e.g. macrocells and small cells. The effect of spatial separation between tiers leads us to model BS locations as a Poisson hole process (PHP). In this model, holes are carved out around the macrocell BSs (MBSs) locations in which small cell BSs (SBSs) cannot be deployed. In this case, the MBSs and the SBSs deployments follow a PPP and a PHP, respectively. Using tools from stochastic geometry, this paper proposes a framework to evaluate the coverage performance in a two-tier mmWave HCN. Since the exact formulation of distance distribution, association probability and coverage probability based on PHP modeling of SBSs is not known, fairly accurate expressions for them are derived and validated by simulation results. It turns out that the analysis based on the proposed model is more accurate for a two-tier HCN than independent PPP-based analysis.