LLM-Enabled LAWNs: Toward Integrated Sensing, Communication, and Control

Low-altitude wireless networks (LAWNs) play an important role in supporting unmanned aerial vehicles (UAVs) and related services. However, achieving seamless integration of sensing, communication, and control remains a core challenge in dynamic environments, where traditional methods fall short in adaptability and efficiency. This paper investigates the integration of large language models (LLMs) and LAWNs to facilitate adaptive integrated sensing, communication, and control performance. We first outline foundational concepts of LAWNs and LLMs, including their main applications and strengths. Next, we examine the benefits and challenges of integrating LLMs and LAWNs. Additionally, we propose a novel LLM-enabled LAWN that dynamically integrates sensing for environmental perception, communication for robust data exchange, and control for precise trajectory management. A case study on urban UAV operations demonstrates clear performance gains: the proposed method reduces total route length by nearly 20% and achieves the highest average data rate among all benchmark planners while maintaining 95% sensing accuracy. These results show that LLM-driven decision support can improve both mission efficiency and communication quality in practical LAWNs.