Although the percentage of people living outside a broadband network has more than halved in recent years, around 10% of the global population does not have access to the Internet. This lack of coverage is particularly concentrated in rural and low-income areas, in which the lack of a cost-effective electricity supply is the main barrier to expanding network coverage. To tackle this problem, this work proposes a theoretical model based on a self-sustainable 5G network architecture in which the mission planning of a swarm of Unmanned Aerial Vehicles (UAVs) is efficiently scheduled to provide cellular coverage over the territory and to reduce the required energy consumption. A Mixed Integer Linear Programming is provided to formalize the problem of minimizing the energy consumption required by the swarm of UAVs that are able to provide coverage by operating at different altitudes. In order to practically solve the problem, a Genetic Algorithm is defined and evaluated over a realistic scenario. Results indicate that a higher granularity in the number of altitudes at which UAVs can provide coverage increases the percentage of territory that is covered, while a small penalty on the energy consumption must be paid compared to the case in which an unique altitude over the ground is considered.

尽管近年来居住在宽带网络之外的人口比例减少了一半以上，但约有10％的全球人口无法使用Internet。这种覆盖不足的情况尤其集中在农村和低收入地区，那里缺乏具有成本效益的电力供应是扩大网络覆盖范围的主要障碍。为了解决这个问题，这项工作提出了一种基于自我可持续发展的5G网络架构的理论模型，其中有效安排了无人飞行器（UAV）群的任务计划，以提供整个区域的蜂窝覆盖范围并减少所需的空间能源消耗。提供了混合整数线性规划来规范化使能够通过在不同高度下运行而提供覆盖范围的无人机群所需的能量消耗最小化的问题。为了实际解决该问题，在现实情况下定义并评估了遗传算法。结果表明，无人机可以提供覆盖的更高海拔高度上的颗粒度会增加覆盖区域的百分比，而与地面上唯一的海拔高度相比，必须付出少量的能耗代价经过考虑的。