The use of unmanned aerial vehicles (UAVS) has been suggested as a potential communications alternative due to their fast implantation, which makes this resource an ideal solution to provide support in scenarios such as natural disasters or intentional attacks that may cause partial or complete disruption of telecommunications services. However, one limitation of this solution is energy autonomy, which affects mission life. With this in mind, our group has developed a new method based on reinforcement learning that aims to reduce the power consumption of UAV missions in disaster scenarios to circumvent the negative effects of wind variations, thus optimizing the timing of the aerial mesh in locations affected by the disruption of fiber-optic-based telecommunications. The method considers the K-means to stagger the position of the resource stations—from which the UAVS launched—within the topology of Stockholm, Sweden. For the UAVS’ locomotion, the Q-learning approach was used to investigate possible actions that the UAVS could take due to urban obstacles randomly distributed in the scenario and due to wind speed. The latter is related to the way the UAVS are arranged during the mission. The numerical results of the simulations have shown that the solution based on reinforcement learning was able to reduce the power consumption by 15.93% compared to the naive solution, which can lead to an increase in the life of UAV missions.

The rapid growth of data traffic due to the demands of new services and applications poses new challenges to the wireless network. Unmanned aerial vehicles (UAVs) can be a solution to support wireless networks during congestion, especially in scenarios where the region has high traffic peaks due to the temporal and spatial flow of users. In this paper, an intelligent machine-learning-based system is proposed to deploy UAV base stations (UAV-BS) to temporarily support the mobile network in regions suffering from the congestion effect caused by the high density of users. The system includes two main steps, the load prediction algorithm (LPA) and the UAV-BSs clustering and positioning algorithm (UCPA). In LPA, the load history generated by the mobile network is used to predict which macrocells are congested. In UCPA, planning is performed to calculate the number of UAV BSs needed based on two strategies: naïve and optimized, in addition to calculating the optimal positioning for each device requested to support the overloaded macrocells. For prediction, we used two models, generalized regression neural networks (GRNN) and random forest, and the results showed that both models were able to make accurate predictions, and the random forest model was better with an accuracy of over 85%. The results showed that the intelligent system significantly reduced the overhead of the affected macrocells, improved the quality of service (QoS), and reduced the probability of blocking users, as well as defined the preventive scheduling for the UAV BSs, which benefited the scheduling and energy efficiency.