Search papers, labs, and topics across Lattice.
This paper introduces an Agentic AI-enhanced, dual spatial-scale framework for optimizing UAV network topology, leveraging exact potential games (EPGs) for both inter-UAV link configuration and UAV deployment. A log-linear learning EPG (L3-EPG) optimizes link configurations, while an approximate gradient EPG (AG-EPG) handles deployment, power allocation, and user association. The framework incorporates a large language model (LLM) to dynamically adjust utility weights based on network characteristics, improving adaptability and reducing the need for manual tuning.
An LLM can automatically tune UAV network optimization, outperforming hand-tuned baselines in energy consumption, latency, and throughput.
Unmanned Aerial Vehicular Networks (UAVNs) are envisioned to provide flexible connectivity, wide-area coverage, and low-latency services in dynamic environments. From an agentic artificial intelligence (Agentic AI) perspective, UAVNs naturally operate as multi-agent systems, where autonomous UAVs act as intelligent agents that coordinate deployment and networking decisions to achieve global performance objectives. However, the strong coupling between discrete link decisions and continuous deployment parameters makes UAVN topology optimization a mixed-integer nonconvex problem, resulting in challenges in scalability, efficiency, and solution consistency under dynamic network conditions. This paper proposes a dual spatial-scale UAVN topology optimization framework based on exact potential games (EPGs), enhanced by Agentic AI. At the large spatial scale, a log-linear learning based EPG (L3-EPG) algorithm is developed to optimize inter-UAV link configurations, enabling sparse yet connected network topologies while reducing redundant links and interference. At the small spatial scale, an approximate gradient based EPG (AG-EPG) algorithm jointly optimizes UAV deployment, transmission power allocation, and ground user (GU) association to improve network throughput and latency. To further enhance adaptability across heterogeneous scenarios, a large language model (LLM) is incorporated as a knowledge-driven decision enhancer to automatically generate utility weights according to network characteristics, alleviating reliance on manual parameter tuning. Simulation results demonstrate that the proposed framework consistently outperforms baseline methods in terms of energy consumption, end-to-end latency, and system throughput.
