Diffusion-OAMP for Joint Image Compression and Wireless Transmission

TL;DR

This paper introduces Diffusion-OAMP, a novel training-free framework combining diffusion models with OAMP for joint image compression and wireless transmission, showing improved performance over classic methods.

Contribution

It proposes a new method that embeds pre-trained diffusion models into the OAMP algorithm for joint image compression and wireless transmission, without requiring training.

Findings

Diffusion-OAMP outperforms classic methods in various compression and noise scenarios.

The framework effectively incorporates multiple generative priors into OAMP.

Experiments demonstrate favorable performance across different compression ratios and noise levels.

Abstract

Joint image compression and wireless transmission remain relatively underexplored compared to generic image restoration, despite its importance in practical communication systems. We formulate this problem under an equivalent linear model, and propose Diffusion-OAMP, a training-free reconstruction framework that embeds a pre-trained diffusion model into the OAMP algorithm. In Diffusion-OAMP, the OAMP linear estimator produces pseudo-AWGN observations, while the diffusion model serves as a nonlinear estimator under an SNR-matching rule. This framework offers a way to incorporate multiple generative priors into OAMP. Experiments with varying compression ratios and noise levels show that Diffusion-OAMP performs favorably against classic methods in the evaluated settings.