Intelligent UAV Base Station Placement with Generative AI and Retrieval-Augmented LLMs

Fixed network infrastructure is often insufficient in addressing dynamic communication demands, particularly in densely populated urban areas. Unmanned Aerial Vehicles (UAVs), specifically UAV-based stations (UAV-BS), offer a flexible and adaptive solution by acting as mobile base stations to alleviate data traffic congestion. This paper introduces a novel framework that leverages generative artificial intelligence (GAI) to optimize the placement of UAV-BS in urban environments. The proposed system integrates GAI agents with LlamaIndex and retrieval-augmented generation (RAG) techniques powered by large language models (LLMs) to enhance decision-making. LlamaIndex serves as an intermediary between user queries and a curated database of UAV-related research, enabling the framework to retrieve and utilize semantically relevant information through RAG. This combination allows the GAI agents to dynamically adapt to real-time data and user requirements, optimizing UAV-BS deployment for maximum efficiency. Simulation results highlight the accuracy and computational efficiency of the proposed approach, demonstrating its superiority in resolving UAV-BS placement challenges compared to traditional methods. This work offers a scalable and intelligent solution for next-generation urban network deployment, paving the way for smart and responsive communication systems.