Indoor Fluid Antenna Systems Enabled by Layout-Specific Modeling and Group Relative Policy Optimization

The fluid antenna system (FAS) revolutionizes wireless communications by employing position-flexible antennas that dynamically optimize channel conditions and mitigate multipath fading. This innovation is particularly valuable in indoor environments, where signal propagation is severely degraded due to structural obstructions and complex multipath reflections. In this paper, we study the channel modeling and joint optimization of antenna positioning, beamforming, and power allocation for indoor FAS. In particular, we propose, for the first time, a layout-specific channel model and a novel group relative policy optimization (GRPO) algorithm for indoor FAS. Compared to the state-of-the-art Sionna model, our approach achieves an $83.3\%$ reduction in computation time with an approximately $3$ dB increase in root-mean-square error (RMSE). When simplified to a two-ray model, our channel model enables a closed-form solution for the optimal antenna position, achieving near-optimal performance. {For the joint optimization problem, the proposed GRPO algorithm outperforms proximal policy optimization (PPO) and other baselines in sum-rate, while requiring only 49.2\% computational resources of PPO, due to its group-based advantage estimation.} Simulation results reveal that increasing either the group size or trajectory length in GRPO does not yield significant improvements in sum-rate, suggesting that these parameters can be selected conservatively without sacrificing performance.