Accurate and Fast Channel Estimation for Fluid Antenna Systems with Diffusion Models

TL;DR

The paper tackles CSI acquisition for high-dimensional fluid antenna systems with limited RF chains. It introduces a diffusion model–based learned prior and posterior sampling (DDRM) to reconstruct the full channel from partial port observations, with a skipped-sampling scheme to speed up online inference. The approach demonstrates superior estimation accuracy and substantial online speedups over compressed sensing baselines, enabling practical deployment in high-dimensional FAS. This work highlights diffusion models as effective priors for spatially correlated antenna channels and offers a scalable path toward real-time 2D FAS channel estimation.

Abstract

Fluid antenna systems (FAS) offer enhanced spatial diversity for next-generation wireless systems. However, acquiring accurate channel state information (CSI) remains challenging due to the large number of reconfigurable ports and the limited availability of radio-frequency (RF) chains -- particularly in high-dimensional FAS scenarios. To address this challenge, we propose an efficient posterior sampling-based channel estimator that leverages a diffusion model (DM) with a simplified U-Net architecture to capture the spatial correlation structure of two-dimensional FAS channels. The DM is initially trained offline in an unsupervised way and then applied online as a learned implicit prior to reconstruct CSI from partial observations via posterior sampling through a denoising diffusion restoration model (DDRM). To accelerate the online inference, we introduce a skipped sampling strategy that updates only a subset of latent variables during the sampling process, thereby reducing the computational cost with minimal accuracy degradation. Simulation results demonstrate that the proposed approach achieves significantly higher estimation accuracy and over 20x speedup compared to state-of-the-art compressed sensing-based methods, highlighting its potential for practical deployment in high-dimensional FAS.

Figures (5)

• Figure 1: An illustration of the point-to-point system considered, where the receiver is equipped with an $N$-port 2D FAS and has $M$ RF-chains.
• Figure 2: An illustration of the denoising network $\boldsymbol{\epsilon}_{\boldsymbol{\theta}}$ used in the DM, implemented using a simplified U-Net structure with sinusoidal time embedding.
• Figure 3: NMSE performance versus SNR.
• Figure 4: NMSE performance versus sampling ratio.
• Figure 5: NMSE performance versus sampling ratio.