Providing wireless services from the sky, using aerial networks (aerial platforms), has recently gained significant attention. Aerial platforms such as (UAVs, also referred to as drones), balloons, and high-altitude/medium-altitude/low-altitude platforms (HAPs/MAPs/LAPs) can play a key role in future wireless networks. For example, a base station (BS) mounted on an aerial platform can boost capacity and/or enhance coverage. This thesis aims to address the challenges accompanied with the use of aerial platforms in wireless communication networks. Specifically, we propose novel frameworks that address various issues related to backhauling/fronthauling a dense deployment of small cells and the recently-envisioned concept of aerial-BSs.

This thesis starts with proposing a novel backhaul/fronthaul network capable of transporting the backhaul/fronthaul traffic between the terrestrial-BSs and the core network. In the proposed network, the aerial platforms act as aerial-hubs that collect/deliver traffic from/to small cells via free-space optics (FSO) links. We show the main limitations of the proposed network and identify proper ways to tackle these limitations.

Aerial platforms can also act as aerial-BSs; therefore, we address the energy-efficient aerial-BS placement problem. The aim is to find the 3D location of the aerial-BS that maximizes the number of covered users using the minimum transmit power. We decouple the aerial-BS deployments in the horizontal and vertical dimensions without any loss of optimality. Next, we investigate the 3D aerial-BS placement that maximizes the number of covered users with different quality-of-service (QoS) requirements. This 3D placement problem is modeled as a multiple circles placement problem. We propose an optimal placement algorithm. A low-complexity algorithm, referred to as the maximal weighted algorithm, is also proposed.

In the last part of the thesis, stochastic geometry-based frameworks are proposed to analyze coverage and rate in two deployment scenarios: (1) a network of aerial-BSs with no underlying terrestrial-BSs, (2) a network of aerial-BSs and terrestrial-BSs. We derive analytical expressions for the conditional Laplace transform of the interference power and the association probabilities. Exact and approximate analytical expressions for the coverage probability and average achievable rate are also derived.