Tractable Performance Analysis for Outdoor and Indoor 3D Small-cell Networks

In order to cope with the exponential growth in wireless capacity demands, network operators will deploy a large number of small cells to improve spatial spectrum reuse. Compared with conventional sparse cellular networks, small-cell networks (SCNs) have much shorter transmission links, and therefore the height difference between base stations (BSs) and users has a significant impact on the network performance with respect to coverage and capacity.
However, most existing works have modelled SCNs on a two-dimensional plane, which may be highly inaccurate. Moreover, the coexisting of small cells and regular macrocells, and the application of emerging technologies such as large antenna arrays and millimeter wave (mmWave) communications have brought new research challenges in the deployment of small cells. To address these challenges, this thesis develops new tractable models using stochastic geometry for three three-dimensional (3D) SCN scenarios: 3D mmWave SCNs, 3D heterogeneous networks (HetNets) and indoor multi-storey SCNs.

Communication in mmWave spectrum is one key enabler to provide high data rates. In the first paper, we propose a 3D system model for outdoor mmWave SCNs, capturing the 3D nature of the deployment environment and characterizing antenna array gains in both elevation and azimuth dimensions. We analytically derive the downlink (DL) coverage probability and area spectral efficiency (ASE). Our results reveal that when other network parameters are fixed, the joint optimization of BS height and BS antenna downtilt can maximize the DL coverage probability and ASE.

In the second paper, we model and analyze an outdoor K-tier 3D HetNet where different tiers have potential different BS density, BS height, transmit power, number of antennas per BS, path loss exponent and cell association bias. Based on the proposed model, the expressions of the DL ergodic rate, ASE and energy efficiency are derived under both the strongest received signal and the closest BS cell-association strategies. We observe that in an ultra-dense HetNet, under both cell-association strategies, SBSs should be deployed at the same height as users’ antennas to achieve high ergodic rate, ASE and energy efficiency.

Finally, in the third paper, we develop an indoor multi-storey SCN model, incorporating the storey height and ceiling penetration loss. We analytically derive the DL coverage probability, spectral efficiency and ASE. Simulation results show that there exist certain values of storey height and BS density that degrade the DL coverage probability. The results can shed new insights into the deployment of indoor small cells and the design of a new multi-storey building from the perspective of enhancing indoor wireless coverage.

The analytical and numerical results presented in this thesis provide guidelines for the planning and deployment of 3D mmWave SCNs, 3D HetNets and indoor multi-storey SCNs, and the proposed analytical frameworks based on stochastic geometry can be extended to the modelling and analysis of other 3D cellular network scenarios.