AI‐Driven Resource Allocation and Beamforming in 6G Terahertz Networks

ABSTRACT The sixth‐generation (6G) wireless networks are envisioned to deliver unprecedented capabilities, including ultra‐high data rates, ultra‐low latency, and massive connectivity. Among the emerging technologies, the terahertz (THz) spectrum is a key enabler for achieving terabit‐per‐second transmission speeds. However, THz communication faces significant obstacles, such as severe path loss, frequent signal blockage, and high channel variability. To address these challenges, this paper introduces an AI‐driven resource allocation and beamforming in 6G terahertz networks (AI‐DRAB‐6G‐THz) that integrates dynamic resource allocation and adaptive beamforming for enhanced performance in 6G THz networks. The framework employs deep reinforcement learning (DRL) to intelligently manage spectrum and power resources in real time, thereby optimizing both spectral efficiency and energy consumption. Simultaneously, a neural network–based beamforming model predicts optimal beam angles and alignment strategies by learning from user mobility patterns and channel state information (CSI). A MATLAB‐based 0–1000 rounds simulation environment, incorporating a realistic THz channel model and environmental constraints, is developed to assess system performance. Evaluation across multiple key metrics, including throughput, latency, spectral efficiency, and beam alignment accuracy, demonstrates that the proposed AI‐driven approach significantly outperforms traditional heuristic methods. Overall, this work underscores the feasibility and effectiveness of integrating AI into the physical layer of 6G systems, paving the way for intelligent, adaptive, and energy‐efficient wireless communication in future 6G networks.