Plug-and-Play AMC: Context Is King in Training-Free, Open-Set Modulation with LLMs

2025-05-06

AI Summary

This work addresses the open-set automatic modulation classification (AMC) problem under three practical constraints: no retraining, no preprocessing, and single-shot inference. We propose the first zero-training, plug-and-play signal–language coupling framework. Methodologically, classical signal features—such as higher-order cumulants—are linguistically encoded into natural language prompts augmented with exemplar contexts, directly prompting open-source large language models (e.g., Llama, Mistral) for one-shot classification. No fine-tuning or channel-specific model adaptation is required. Our key contribution lies in achieving zero-parameter coupling between signal processing and LLMs: contextual prompting enables implicit modeling of communication-domain priors within the LLM, yielding inherent noise robustness, cross-modulation generalization, and preprocessing-free operation. The framework achieves state-of-the-art performance across diverse modulation schemes and SNR levels, significantly reducing deployment overhead. Code and experimental results are publicly released and empirically validated.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Automatic Modulation Classification (AMC) is critical for efficient spectrum management and robust wireless communications. However, AMC remains challenging due to the complex interplay of signal interference and noise. In this work, we propose an innovative framework that integrates traditional signal processing techniques with Large-Language Models (LLMs) to address AMC. Our approach leverages higher-order statistics and cumulant estimation to convert quantitative signal features into structured natural language prompts. By incorporating exemplar contexts into these prompts, our method exploits the LLM's inherent familiarity with classical signal processing, enabling effective one-shot classification without additional training or preprocessing (e.g., denoising). Experimental evaluations on synthetically generated datasets, spanning both noiseless and noisy conditions, demonstrate that our framework achieves competitive performance across diverse modulation schemes and Signal-to-Noise Ratios (SNRs). Moreover, our approach paves the way for robust foundation models in wireless communications across varying channel conditions, significantly reducing the expense associated with developing channel-specific models. This work lays the foundation for scalable, interpretable, and versatile signal classification systems in next-generation wireless networks. The source code is available at https://github.com/RU-SIT/context-is-king

Problem

• Enhancing Automatic Modulation Classification without training
• Integrating signal processing with LLMs for open-set modulation
• Improving robustness in wireless communications across varying conditions

Innovation

• Integrates signal processing with LLMs for AMC
• Uses natural language prompts from signal features
• Enables one-shot classification without training