DRL-Based Joint Beamforming and Surface Morphing for FIM-Enabled ISAC Systems

Abstract

In this paper, we investigate the sum rate maximization problem for integrated sensing and communication (ISAC) systems enabled by a flexible intelligent metasurface (FIM) deployed at the base station. By enabling element movement through surface morphing, the FIM introduces additional spatial degrees of freedom that can reshape the wireless propagation environment. To exploit this capability, we jointly optimize the transmit digital beamforming at the base station and the FIM surface configuration, characterized by the positions of its radiating elements, subject to transmit power constraints and minimum sensing beam gain requirements. Since the resulting problem is highly non-convex due to the strong coupling between the beamforming vectors and the element positions, a deep reinforcement learning (DRL)-based algorithm is proposed to efficiently obtain high-quality solutions. Numerical results demonstrate that the proposed framework significantly improves the achievable sum rate while satisfying the sensing performance constraints.