Latency and power optimization in terahertz UAV-assisted vehicular networks across diverse atmospheric profile conditions

Terahertz (THz) communication has emerged as a key technology for high-speed wireless networks, particularly in scenarios where conventional frequency bands fail to meet growing data demands. With its potential for ultra-low latency, broad bandwidth, and robust connectivity, THz communication offers a suitable infrastructure for intelligent transportation systems and autonomous vehicles, especially within Vehicle-to-Everything (V2X) and Unmanned Aerial Vehicle (UAV) communication networks. This study aims to optimize THz communication between UAVs and ground vehicles under varying atmospheric conditions. Specifically, an artificial intelligence (AI)-based scheme is proposed to simultaneously minimize latency and transmission power while maintaining a sufficient signal-to-noise ratio (SNR) for successful communication. The proposed method integrates a dual-objective Particle Swarm Optimization (PSO) algorithm with the Line-by-Line Radiative Transfer Model (LBLRTM), which accurately models atmospheric absorption characteristics. Designed for critical scenarios such as emergency response operations, the scheme dynamically determines UAV positions and transmission powers to ensure both energy efficiency and low-latency communication. Simulation results demonstrate that the proposed approach achieves sufficient SNR levels and low latency across all atmospheric models. These findings highlight the potential of the AI-based approach to enhance energy efficiency and ensure sustainable connectivity in THz-enabled networks for time-sensitive applications.