Greedy and Transformer-Based Multi-Port Selection for Slow Fluid Antenna Multiple Access
This addresses a computational bottleneck in wireless communication systems, offering incremental improvements over existing methods.
The paper tackles the port-selection problem in fluid antenna multiple access systems by proposing two strategies: a greedy method with swap refinement that outperforms state-of-the-art schemes in spectral efficiency, and a Transformer-based neural network that approaches this performance at lower computational cost.
We address the port-selection problem in fluid antenna multiple access (FAMA) systems with multi-port fluid antenna (FA) receivers. Existing methods either achieve near-optimal spectral efficiency (SE) at prohibitive computational cost or sacrifice significant performance for lower complexity. We propose two complementary strategies: (i) GFwd+S, a greedy forward-selection method with swap refinement that consistently outperforms state-of-the-art reference schemes in terms of SE, and (ii) a Transformer-based neural network trained via imitation learning followed by a Reinforce policy-gradient stage, which approaches GFwd+S performance at lower computational cost.