Dependability Theory-Based Statistical QoS Provisioning of Fluid Antenna Systems

Fluid antenna systems (FAS) have recently emerged as a promising technology for next-generation wireless networks, offering real-time spatial reconfiguration to enhance reliability, throughput, and energy efficiency. Nevertheless, existing studies often overlook the temporal dynamics of channel fading and their implications for mission-critical operations. In this paper, we propose a dependability-theoretic framework for statistical quality-of-service (QoS) provisioning of FAS under finite blocklength (FBL) constraints. Specifically, we derive new closed-form expressions for the level-crossing rate (LCR) and average fade duration (AFD) of an N -port FAS over Nakagami- m fading channels. Leveraging these second-order statistics, we introduce two key dependability metrics, such as mission reliability and mean time-to-first-failure (MTTFF), to quantify the probability of uninterrupted operation over a defined mission duration. We further extend the classical effective capacity (EC) concept to incorporate mission reliability in the FBL regime, yielding a new performance metric, mission EC (mEC). To capture energy efficiency under bursty traffic and latency constraints, we also develop the mission effective energy efficiency (mEEE) metric and formulate its maximisation as a non-convex fractional optimisation problem. This problem is then solved via a modified Dinkelbach’s method with an embedded line search. Extensive simulations uncover critical trade-offs among port count, QoS exponent, signal-to-noise ratio, and mission duration, offering insights for the design of ultra-reliable, low-latency, and energy-efficient industrial Internet of Things (IIoT) systems. © 2002-2012 IEEE.