A Gaussian Process Framework for Outage Analysis in Continuous-Aperture Fluid Antenna Systems

TL;DR

This paper introduces a Gaussian process-based analytical framework for outage probability analysis in fluid antenna systems, providing closed-form expressions, bounds, and extreme value theory insights that improve understanding of system reliability.

Contribution

It develops a novel Gaussian kernel-based approach for outage analysis in fluid antenna systems, including closed-form expressions, bounds, and extreme value theory, applicable to continuous apertures.

Findings

Gaussian approximation yields less than 10% outage error for W ≤ 2

Outage probability depends mainly on aperture W and threshold x

Convergence observed with approximately 10W ports

Abstract

This paper develops a comprehensive analytical framework for the outage probability of fluid antenna system (FAS)-aided communications by modeling the antenna as a continuous aperture and approximating the Jakes (Bessel) spatial correlation with a Gaussian kernel $\rho_G(\delta) = e^{-\pi^2\delta^2}$. Three complementary analytical strategies are pursued. First, the Karhunen--Lo\`{e}ve (KL) expansion under the Gaussian kernel is derived, yielding closed-form outage expressions for the rank-1 and rank-2 truncations and a Gauss--Hermite formula for arbitrary rank~$K$, with effective degrees of freedom $K_{\mathrm{eff}}^G \approx \pi\sqrt{2}\, W$. Second, rigorous two-sided outage bounds are established via Slepian's inequality and the Gaussian comparison theorem: by sandwiching the true correlation between equi-correlated models with $\rho_{\min}$ and $\rho_{\max}$, closed-form upper and lower bounds that avoid the optimistic bias of block-correlation models are obtained. Third, a continuous-aperture extreme value theory is developed using the Adler--Taylor expected Euler characteristic method and Piterbarg's theorem. The resulting outage expression $P_{\mathrm{out}} \approx 1 - e^{-x}(1 + \pi\sqrt{2}\, W\, x)$ depends only on the aperture~$W$ and threshold~$x$, is independent of the port count~$N$, and is identical for the Jakes and Gaussian models since both share the second spectral moment $\lambda_2 = 2\pi^2$. A Pickands-constant refinement for the deep-outage regime and a threshold-dependent effective diversity $N_{\mathrm{eff}} \approx 1 + \pi\sqrt{2}\, W\, x$ are further derived. Numerical results confirm that the Gaussian approximation incurs less than 10\% relative outage error for $W \leq 2$ and that the continuous-aperture formula converges with as few as $N \approx 10W$ ports.