Nuts and Bolts of a Realistic Stochastic Geometric Analysis of mmWave HetNets: Hardware Impairments and Channel Aging

Motivated by heterogeneous networks (HetNets) design in improving coverage and by millimeter-wave (mmWave) transmission offering an abundance of extra spectrum, we present a general analytical framework shedding light to the downlink of realistic mmWave HetNets consisting of $K$ tiers of randomly located base stations (BSs). Specifically, we model, by virtue of stochastic geometry tools, the $\mathit{\text{multi-tier}}$ \textit{multi-user multiple-input multiple-output (MU-MIMO) mmWave} network degraded by the inevitable $\mathit{\text{residual}}\mathit{\text{additive}}\mathit{\text{transceiver}}\mathit{\text{hardware}}\mathit{\text{impairments}}\mathit{\text{(RATHIs)}}$ and $\mathit{\text{channel}}\mathit{\text{aging}}$. Given this setting, we derive the coverage probability and the area spectral efficiency (ASE), and we subsequently evaluate the impact of residual transceiver hardware impairments (RTHIs) and channel aging on these metrics. Different path-loss laws for line-of-sight (LOS) and non-line-of-sight (NLOS) are accounted for the analysis, which are among the distinguishing features of mmWave systems. Among the findings, we show that the RATHIs have a meaningful impact at the high signal-to-noise ratio (SNR) regime, while the transmit additive distortion degrades further than the receive distortion the system performance. Moreover, serving fewer users proves to be preferable, and the more directive the mmWaves are, the higher the ASE becomes.