ModFus-DM: Explore the Representation in Modulated Signal Diffusion Generated Models

TL;DR

ModFus-DM introduces an unsupervised diffusion-based framework for automatic modulation classification that effectively learns robust signal representations, outperforming existing methods especially with limited labels and challenging conditions.

Contribution

The paper proposes ModFus-DM, a novel diffusion model-based approach for AMC that captures structural features and adaptively fuses multi-scale diffusion features for improved classification.

Findings

Achieves over 88.27% accuracy with only 10 labeled signals per class.

Outperforms existing methods in limited-label and distribution shift scenarios.

Effective in recognizing variable-length signals and channel fading conditions.

Abstract

Automatic modulation classification (AMC) is essential for wireless communication systems in both military and civilian applications. However, existing deep learning-based AMC methods often require large labeled signals and struggle with non-fixed signal lengths, distribution shifts, and limited labeled signals. To address these challenges, we propose a modulation-driven feature fusion via diffusion model (ModFus-DM), a novel unsupervised AMC framework that leverages the generative capacity of diffusion models for robust modulation representation learning. We design a modulated signal diffusion generation model (MSDGM) to implicitly capture structural and semantic information through a progressive denoising process. Additionally, we propose the diffusion-aware feature fusion (DAFFus) module, which adaptively aggregates multi-scale diffusion features to enhance discriminative representation. Extensive experiments on RML2016.10A, RML2016.10B, RML2018.01A and RML2022 datasets demonstrate that ModFus-DM significantly outperforms existing methods in various challenging scenarios, such as limited-label settings, distribution shifts, variable-length signal recognition and channel fading scenarios. Notably, ModFus-DM achieves over 88.27% accuracy in 24-type recognition tasks at SNR $\geq $ 12dB with only 10 labeled signals per type.