RF-Analyzer: Can Vision-Language Models Learn RF Understanding from Synthetic Data?

TL;DR

This paper explores whether vision-language models trained solely on synthetic spectrogram data can effectively interpret real RF signals, demonstrating promising generalization but with notable limitations in low-SNR conditions.

Contribution

It introduces RF-Analyzer, a platform for evaluating VLMs on live RF signals, and establishes a benchmark showing synthetic data can enable transferable RF understanding.

Findings

VLMs trained on synthetic data can extract physical RF attributes from real signals.

Synthetic training enables some generalization but struggles with low-SNR and unseen conditions.

The benchmark framework quantifies signal understanding and grounding in RF interpretation.

Abstract

Understanding the wireless spectrum is a fundamen- tal requirement for intelligent communication systems, however, interpreting spectrograms requires extracting multiple physical attributes and reasoning about signal structure, which is a capability that is not achieved by traditional ML approaches. Recent advances in vision-language models (VLMs) demonstrated the possibility of learning such interpretation capabilities directly from data. This paper investigates whether VLMs can learn this capability from synthetic data alone, and more importantly, whether such learned representations generalize to real over-the- air RF environments. To address this question, we introduce RF-Analyzer, an SDR-to-AI analysis platform that integrates live spectrum captures associated with the corresponding VLM- based interpretation, enabling direct evaluation of VLMs outputs on live over-the-air signals. Using this platform, we assess a model trained exclusively on synthetic spectrogram data with general-purpose baselines. To enable systematic analysis, we establish a benchmark framework comprising three metrics, Physical Attribute Extraction Score (PAES), Prompt Leakage Rate (PLR), and hallucination count, to assess signal understanding and grounding. The obtained results demonstrate that VLMs trained on synthetic spectrogram data can generalize to real RF environments, particularly for extracting physical signal attributes such as spectral occupancy, temporal behavior, and SNR. This indicates that synthetic data is sufficient for learning transferable representations of RF signal structure. However, this generalization is limited due to the fact that synthetic training does not provide reliable semantic grounding without contextual priors. In particular, generalization breaks under conditions that are not covered in the synthetic distribution, particularly low-SNR regimes