Decision Transformer improves UAV device communication with intelligent surfaces
Decision Transformer for UAV-Mounted RIS-Assisted Dynamic D2D Communications
Artificial IntelligenceRobotics
Summary
Communicating between devices directly, like phones linked without cell towers, gets tricky when signals change or move around. The authors studied how to use drones carrying special reflective panels called reconfigurable intelligent surfaces (RIS) to boost these signals. They trained an AI method called a Decision Transformer to figure out the best drone paths and panel settings to make communication as fast as possible. Their approach works well across different situations without retraining and can learn quickly with less trial and error compared to other AI methods.
What this means in practice
- •For wireless network engineers: Optimize drone paths and reflective surface settings to improve direct device communication in variable environments.
- •For drone operation teams: Deploy UAV-mounted intelligent surfaces with AI guidance to dynamically enhance device connectivity with minimal setup time.
Authors
Yaxuan Liu
Abstract
This paper studies unmanned aerial vehicle (UAV)-mouted reconfigurable intelligent surface (RIS)-assisted device-to-device (D2D) communication with stochastic link activation. It models UAV motion and attitude, time-varying Rician angles, and angle-dependent RIS reflection. A joint optimization of UAV trajectory, attitude, and RIS phases is formulated to maximize average sum rate under mobility, energy, and hardware constraints. The problem is addressed using deep reinforcement learning and a Decision Transformer trained on expert trajectories from multiple scenarios. Results demonstrate effective cross-scenario generalization, with zero-shot transfer outperforming direct DRL transfer and online fine-tuning achieving competitive performance with fewer interactions.