Analysis and Optimization for Low-Latency Communications in Slow Fluid Antenna Multiple Access Systems

The fluid antenna system (FAS) is a reconfigurable antenna technology that enhances wireless communications by adapting to time-varying channel conditions. However, existing FAS research focuses on physical-layer performance, neglecting link-layer quality-of-service (QoS) guarantees like latency/delay. This paper develops a cross-layer model to investigate the low-latency performance of a slow fluid antenna multiple access (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">s</i>-FAMA) system in a multi-user downlink scenario, employing the effective capacity (EC) framework. We first derive approximate expressions for the total EC in a multi-user <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">s</i>-FAMA system under both complex and simplified channel models. To validate them, we propose a quasi-Monte Carlo (QMC) method to compute multidimensional integrals, overcoming the limitations of conventional numerical methods and solving a class of multidimensional integral numerical simulation problems for large-scale FAMA systems. Then, we formulate an optimization problem to maximize the total EC while satisfying the total power constraint and each user’s minimum EC requirement. To jointly optimize port selection and power allocation, an iterative algorithm based on alternating optimization (AO) and quadratic transform (QT) is proposed to solve the non-convex problem. Simulation results validate our approximations and show that our joint port selection and power allocation scheme outperforms the conventional baseline algorithms, confirming our proposed algorithm’s effectiveness and FAS’s superiority in ensuring the QoS in wireless communications.