Browse Teach Curate

Deep Reinforcement Learning for Fluid-Antenna Positioning in ISAC Systems

How to jointly optimize transmit beamforming and continuous fluid-antenna positions in a base station that both communicates and performs multi-target radar sensing. The highly non-convex joint problem is split by block coordinate descent, with the antenna-position subproblem solved by a deep deterministic policy gradient agent that outputs continuous positions directly, enabling scalable real-time balancing of sensing and communication performance.

KITT Text

Towards Intelligent Antenna Positioning: Leveraging DRL for FAS-Aided ISAC Systems Shunxing Yang

Proposes a joint optimization framework for antenna positioning in a fluid antenna system (FAS) used for integrated sensing and communication (ISAC), where a dual-functional base station simultaneous…