Farre V., Vega-Senchez J., Estrada J., Riofrio D., Peralvo J.A.V., Chatzinotas S., Olivo E.E.B.
2025 Workshop on Communication Networks and Power Systems Wcnps 2025, 2025
Fluid Antenna Systems (FAS) promise significant gains by repositioning antenna ports, but network-level deployment is critically constrained by the often-neglected costs of switching latency, pilot overhead, and control signaling. Link-level studies that maximize ergodic rate often assume these costs are negligible, leading to non-viable solutions. To bridge this gap, we propose a novel, cost-aware two-timescale framework that optimizes for net utility, balancing network-wide Area Spectral Efficiency (ASE) against the practical reconfiguration costs. We formulate a scalable optimization problem based on statistical CSI and a probabilistic port-selection surrogate, which is solved efficiently via projected-gradient methods. We then propose three specialized strategies-Interference-Aware (IADP), Load-Balanced (LBSD), and Multi-Objective (MOPO)-that optimize this new utility. System-level simulations in ultra-dense B5G scenarios show our framework achieves substantial gains in net ASE and user throughput, while the explicit cost penalties provably control switching activity and pilot burden, unlike cost-ignorant baselines. Our results provide the first actionable guidelines for the practical, cost-effective deployment of FAS in 6G networks.
