Distributed UAV Swarm for Device-Free Integrated Sensing and Communication Relying on Multi-Agent Reinforcement Learning

Abstract

Sensing and communication sharing the same hardware and frequency band has become a key technology for future 5G/6G systems to achieve high-performance integrated sensing and communication (ISAC), which has attracted a great deal of attention in both academia and industry. In this letter, a device-free ISAC system based on distributed unmanned aerial vehicles (UAVs) is designed to realize the sensing of vehicles on urban roads and transmit information to base stations through wireless communication. In order to realize autonomous decision-making of UAVs and solve the problem of stochastic optimization and multi-index tradeoffs caused by a time-varying environment, we propose an algorithm based on multi-agent reinforcement learning (MARL). The simulation results verify that the proposed scheme can significantly improve the sensing performance of the UAVs and ensure communication quality with base stations. Besides, we realize the sensing fairness for users, the flight safety of UAVs, and can save energy through hyperparameter tuning.