Optimizing Energy Efficiency in Multi-UAV IoT Networks Using Deep Q-Networks (DQN)

This paper focuses on optimizing energy efficiency in a multi-UAV network using a Deep Q Network (DQN) approach, with an emphasis on the implications for network performance in IoT-enabled smart cities. The optimization targets energy efficiency, considering different graph densities (GD) (1, 0.5, and 0.2) across 30 UAVs. Through simulation results, we demonstrate that improving energy efficiency leads to enhanced network performance, including reduced latency. Specifically, energy efficiency reached up to 3.5×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">9</sup> bps/W at a graph density of 1, highlighting the relationship between energy optimization and system responsiveness. By minimizing transmit power, data rates improve, which subsequently reduces latency by approximately 30%. These findings underscore the importance of energy-efficient strategies in future 6G-enabled UAV networks, where enhanced network performance and reduced latency are crucial for supporting smart city applications and IoT connectivity.