RFPrompt: Prompt-Based Expert Adaptation of the Large Wireless Model for Modulation Classification

2026-05-04

AI Summary

This work addresses the limited robustness of automatic modulation classification in real-world wireless environments, where hardware impairments and unknown propagation conditions induce distribution shifts. To tackle this challenge, the authors propose RFPrompt, a novel framework that introduces prompt learning into large wireless models (LWMs) for the first time. RFPrompt adapts downstream tasks in a parameter-efficient manner by injecting learnable deep prompt tokens into a frozen backbone network and integrating a mixture-of-experts architecture. Experimental results demonstrate that RFPrompt significantly enhances generalization under both standard and out-of-distribution settings, achieving remarkable performance on real over-the-air IQ data and in low-supervision scenarios. Notably, it attains strong robustness with only a minimal number of trainable parameters.

Abstract

Automatic modulation classification (AMC) in real-world deployments demands robustness to distribution shifts arising from hardware impairments, unseen propagation environments, and recording conditions never encountered during training. Although wireless foundation models offer a promising starting point for robust RF representation learning, an important open question is how to adapt them efficiently to out-of-distribution (OOD) downstream tasks without overwriting the structure learned during large-scale pre-training. In this paper, we investigate prompt-based adaptation as a general mechanism for OOD transfer in wireless foundation models. We propose RFPrompt, a parameter-efficient framework that introduces learnable deep prompt tokens while keeping the pretrained backbone frozen, enabling task-specific adaptation with minimal trainable parameters. We instantiate and evaluate this approach on the Large Wireless Model (LWM), a mixture-of-experts wireless foundation model, and study its behavior under both standard and OOD modulation-classification settings. Results show that prompt-based adaptation consistently improves robustness under distribution shift and limited supervision, particularly on real-world over-the-air IQ data, while preserving strong parameter efficiency. These findings suggest that prompt learning is a practical and effective strategy for adapting wireless foundation models to challenging downstream RF environments.

Problem

• automatic modulation classification
• distribution shift
• out-of-distribution adaptation
• wireless foundation models
• robustness

Innovation

• prompt-based adaptation
• parameter-efficient learning
• wireless foundation model
• out-of-distribution robustness
• modulation classification