The recent but remarkable emergence of unmanned aerial vehicles (UAVs), which will soon occupy most of our service systems, is key to the expansion of new communication technologies and wireless networks. These machines can be used as flying base stations to extend and/or improve the coverage of the land mobile communication system in order to respond effectively to the ever-increasing demand for traffic. However, determining the best positions for these vehicles during their deployment remains a real and major challenge to achieve satisfactory results from them.
In this study we aim to propose a new 3D deployment scheme of a multi-UAVs network in an area covered by a cluttered or damaged base station (BS), integrating machine learning to maximize the number of users served while maximizing operator benefits.
First, we propose an algorithm to determine the optimal number of UAVs to be deployed to serve all users, considering their varied service requirements. We then determine the optimal locations of these aerial vehicles using a partitioning and positioning algorithm that integrates the auto-encoders to optimize the locations, before finding their altitude.
The simulation of our algorithm is performed under MATLAB. The numerical results show that our proposed Deep K-means approach can serve a very large percentage of users based on the number of UAVs used while maximizing operator profit.
| Date | 22 Feb 2021 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Michel Kadoch (Supervisor) |
|---|
Koffi, A. J. H. (Author),
Kadoch (Supervisor),
22 Feb 2021Student thesis: Master's thesis › Master in Engineering: Engineering