Diffusion-based Generative Prior for Low-Complexity MIMO Channel Estimation

TL;DR

This letter proposes a novel channel estimator based on diffusion models, one of the currently top-rated generative models, with provable convergence to the mean square error (MSE)-optimal estimator, designed to learn the channel distribution in the sparse angular domain.

Abstract

This work proposes a novel channel estimator based on diffusion models (DMs), one of the currently top-rated generative models. Contrary to related works utilizing generative priors, a lightweight convolutional neural network (CNN) with positional embedding of the signal-to-noise ratio (SNR) information is designed by learning the channel distribution in the sparse angular domain. Combined with an estimation strategy that avoids stochastic resampling and truncates reverse diffusion steps that account for lower SNR than the given pilot observation, the resulting DM estimator has both low complexity and memory overhead. Numerical results exhibit better performance than state-of-the-art channel estimators utilizing generative priors.

Figures (5)

• Figure 1: dm architecture with a lightweight cnn and positional embedding of the snr information. The nn parameters are shared across all timesteps.
• Figure 2: mse performance for the 3gpp channel model with three propagation clusters and $T=100$dm timesteps.
• Figure 3: mse performance for the QuaDRiGa channel model and $T=100$dm timesteps.
• Figure 4: mse performance over the total number of dm timesteps $T$ for the QuaDRiGa model.
• Figure 5: mse performance of the intermediate estimates of the dm's timesteps for $T=300$ (solid) and $T=100$ (dashed) for the QuaDRiGa model.