Researchers Bach Hung Luu, Sinh Cong Lam, and Nam Hoang Nguyen from the University of Technology Sydney have proposed a novel approach to improve data rates for cell-edge users in cellular networks, a persistent challenge in the energy and telecommunications sectors. Their work, published in the IEEE Internet of Things Journal, leverages Unmanned Aerial Vehicles (UAVs) and deep learning to optimize power control and enhance network performance.
Cell-edge users (CEUs) often experience lower data rates due to their distance from base stations and physical obstructions. Traditional solutions either deploy UAVs for all users or forgo UAV assistance altogether. The researchers propose a more nuanced approach, using a distance-based criterion to determine which users receive UAV relay assistance. This method ensures that only users beyond a certain reference distance benefit from UAV support, optimizing resource allocation.
Each UAV in this system acts as an amplify-and-forward relay, enabling assisted users to receive signals from both the base station and the UAV simultaneously. This setup achieves diversity gain, improving data rates for CEUs. To optimize transmission power allocation across base stations, the researchers employ a Deep Q-Network (DQN) learning framework. This deep learning approach learns power control policies without requiring accurate channel models, making it adaptable and efficient.
Simulation results demonstrate the effectiveness of the proposed approach. At an optimal reference distance of 400 meters, the system achieves a peak average rate of 2.28 bps/Hz. This represents a 3.6% improvement compared to networks without UAV assistance and a 0.9% improvement compared to networks where all users receive UAV support. The study also highlights the importance of UAV altitude and reference distance in system performance, with lower altitudes providing better results.
For the energy sector, this research offers practical applications in optimizing cellular network performance, particularly in remote or hard-to-reach areas. By improving data rates for CEUs, the approach can enhance connectivity and support the deployment of smart grid technologies, remote monitoring, and other energy-related applications. The use of UAVs and deep learning also aligns with the growing trend of integrating advanced technologies into energy infrastructure to improve efficiency and reliability.
In summary, the researchers have developed an innovative solution to address the performance gap between cell-edge and cell-center users. Their approach combines UAV-assisted communication with deep learning-driven power control, offering significant improvements in data rates and network performance. This research not only advances the field of telecommunications but also has valuable implications for the energy sector, particularly in enhancing connectivity and supporting smart grid technologies.
This article is based on research available at arXiv.