3D Gaussian Radiation Field Modeling for Integrated RIS-FAS Systems: Analysis and Optimization

TL;DR

This paper addresses real-time optimization in integrated RIS-FAS systems under fast fading by introducing a 3D Gaussian Radiation Field (3DGRF) model that represents the electromagnetic energy distribution with explicit Gaussian primitives. A differentiable projection/splatting pipeline renders spatial spectra from limited observations, enabling a Field-Driven Alternating Optimization (FAO) that jointly updates FAS positions and RIS phase shifts using gradient-based and evolutionary methods, without relying on extensive CSI. Key contributions include the SRN for geometry-aware Gaussian parameters, a differentiable 3DGRF reconstruction framework, and a complexity-conscious FAO algorithm with demonstrated improvements in convergence speed, spectral efficiency, and latency over traditional CSI-driven methods. The approach offers a physically interpretable, scalable, and real-time solution for field-domain control in 6G RIS-FAS networks, with practical implications for dynamic environments and multiuser settings.

Abstract

The integration of reconfigurable intelligent surfaces (RIS) and fluid antenna systems (FAS) has attracted considerable attention due to its tremendous potential in enhancing wireless communication performance. However, under fast-fading channel conditions, rapidly and effectively performing joint optimization of the antenna positions in an FAS system and the RIS phase configuration remains a critical challenge. Traditional optimization methods typically rely on complex iterative computations, thus making it challenging to obtain optimal solutions in real time within dynamic channel environments. To address this issue, this paper introduces a field information-driven optimization method based on three-dimensional Gaussian radiation-field modeling for real-time optimization of integrated FAS-RIS systems. In the proposed approach, obstacles are treated as virtual transmitters and, by separately learning the amplitude and phase variations, the model can quickly generate high-precision channel information based on the transmitter's position. This design eliminates the need for extensive pilot overhead and cumbersome computations. On this framework, an alternating optimization scheme is presented to jointly optimize the FAS position and the RIS phase configuration. Simulation results demonstrate that the proposed method significantly outperforms existing approaches in terms of spectrum prediction accuracy, convergence speed, and minimum achievable rate, validating its effectiveness and practicality in fast-fading scenarios.

Figures (10)

• Figure 1: The considered system model of integrated RIS–FAS–assisted uplink multi-user communications.
• Figure 2: Overall framework of the proposed 3DGRF–based optimization. (1) The environment setup includes BS, RIS, and FAS. (2) The scenario representation network encodes geometric and transmitter information into latent radiation-field features. (3) The projection model represents the electromagnetic field using differentiable 3DGS primitives. (4) The electromagnetic splatting process reconstructs the continuous radiation field and generates spatial power distributions. (5) The reconstructed field produces spatial spectrums for subsequent system optimization. (6) The FAO iteratively updates FAS positions and RIS phases based on 3DGRF, achieving field-based control.
• Figure 3: (a)Laboratory environment and LiDAR point-cloud representation for radiation-field reconstruction, where the base station (BS) and multiple users $U_k,U_1,U_j$ are marked. 3D LiDAR point clouds of the laboratory environment were used to initialize the Gaussian primitives for the proposed 3DGS-based field model. The overall experimental scene, where the transmitter (TX) can be placed at arbitrary positions within the environment, and the receiver (RX) equipped with a $4\times4$ antenna array is fixed at the corner of the room.
• Figure 4: (a) Synthesized spatial power spectrum of the RIS–FAS system before optimization, where the integration of RIS reflection and FAS reception forms a preliminary directional pattern. (b) Spectrum after field-driven optimization, showing stronger beam focusing and reduced sidelobes, which demonstrates the effectiveness of the proposed joint configuration of RIS phase and FAS position.
• Figure 5: Minimum achievable rate $R$ comparison between the proposed field-driven optimization and the traditional optimization.

Theorems & Definitions (2)

• proof
• proof