Abstract : Integrated Sensing and Communication (ISAC) has attracted interest as a potential technology for 6G networks because it efficiently combines sensing and communication functions while utilizing shared spectrum resources. ISAC systems use reconfigurable intelligent surfaces (RISs) to dynamically control propagation environments, improving signal quality and coverage in challenging environments for unmanned aerial vehicle (UAV) networks.In this study, we propose a novel solution for joint beamforming in RIS-assisted ISAC systems within UAV networks. By leveraging a deep reinforcement learning (DRL) framework, we aim to optimize beamforming at both the ISAC base station and the RIS mounted on a UAV. The proposed solution maximizes the secrecy rate while ensuring radar detection requirements are met, addressing the challenges posed by non-convex optimization problems. The simulation results demonstrate that deploying RIS within ISAC systems significantly enhances system performance, particularly in termsof secure communication and radar detection, even in dynamic environments such as UAV networks. The proposed solution shows considerable improvements in secrecy rate and adaptability under varying conditions, underscoring the potential of RIS-assisted ISAC for future 6G networks.
Index terms : DRL , ISAC , joint beamforming , RIS , UAV networks