Diffusion Models-Aided Uplink Channel Estimation for RIS-Assisted Systems

2025-06-09

AI Summary

To address the low channel estimation accuracy and poor SNR robustness in RIS-assisted wireless communication systems, this paper proposes a deterministic diffusion model-based channel estimation method. Our approach formulates channel estimation as a denoising inverse process and introduces a deterministic reverse sampling strategy to eliminate the stochasticity inherent in conventional diffusion models. We further design a step-alignment mechanism to enhance noise suppression capability and reconstruction consistency. Additionally, we develop a lightweight U-Net architecture that achieves near-baseline performance with only 6.59% of the parameter count. Experimental results demonstrate that, at SNR = 0 dB, the proposed method reduces normalized mean square error (NMSE) by 13.5 dB compared to baseline methods, significantly improving estimation reliability under low-SNR conditions and enhancing practical deployability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

This letter proposes a channel estimation method for reconfigurable intelligent surface (RIS)-assisted systems through a novel diffusion model (DM) framework. We reformulate the channel estimation problem as a denoising process, which aligns with the reverse process of the DM. To overcome the inherent randomness in the reverse process of conventional DM approaches, we adopt a deterministic sampling strategy with a step alignment mechanism that ensures the accuracy of channel estimation while adapting to different signal-to-noise ratio (SNR). Furthermore, to reduce the number of parameters of the U-Net, we meticulously design a lightweight network that achieves comparable performance, thereby enhancing the practicality of our proposed method. Extensive simulations demonstrate superior performance over a wide range of SNRs compared to baselines. For instance, the proposed method achieves performance improvements of up to 13.5 dB in normalized mean square error (NMSE) at SNR = 0 dB. Notably, the proposed lightweight network exhibits almost no performance loss compared to the original U-Net, while requiring only 6.59% of its parameters.

Problem

• Proposes DM framework for RIS-assisted channel estimation
• Uses deterministic sampling to improve SNR adaptation
• Designs lightweight U-Net to reduce parameters

Innovation

• Uses diffusion model for channel estimation
• Implements deterministic sampling strategy
• Designs lightweight U-Net for efficiency