AIMC-Spec: A Benchmark Dataset for Automatic Intrapulse Modulation Classification Under Variable Noise Conditions

2026-01-13

IEEE Access

AI Summary

This study addresses the limited progress in automatic intra-pulse modulation classification (AIMC) under noisy conditions, primarily hindered by the absence of standardized datasets. To bridge this gap, the authors introduce AIMC-Spec, the first synthetic benchmark dataset encompassing 33 modulation types across 13 signal-to-noise ratio (SNR) levels, with a unified time-frequency spectrogram input format. The work systematically evaluates the classification performance of lightweight CNNs, denoising networks, and Transformers on this dataset. Experimental results reveal that frequency-modulated signals exhibit greater robustness at low SNRs and demonstrate that both modulation type and network architecture significantly influence classification accuracy. By providing a reproducible and comprehensive benchmark, this work establishes a foundation for standardized evaluation and future advancements in AIMC research.

Technology Category

• Machine Learning: Multimodal Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Multi-modal Vision

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

A lack of standardized datasets has long hindered progress in automatic intrapulse modulation classification (AIMC)—a critical task in radar signal analysis for electronic support systems, particularly under noisy or degraded conditions. AIMC seeks to identify the modulation type embedded within a single radar pulse from its complex in-phase and quadrature (I/Q) representation, enabling automated interpretation of intrapulse structure. This paper introduces AIMC-Spec, a comprehensive synthetic dataset for spectrogram-based image classification, encompassing 33 modulation types across 13 signal-to-noise ratio (SNR) levels. To benchmark AIMC-Spec, five representative deep learning algorithms—ranging from lightweight CNNs and denoising architectures to transformer-based networks—were re-implemented and evaluated under a unified input format. The results reveal significant performance variation, with frequency-modulated (FM) signals classified more reliably than phase or hybrid types, particularly at low SNRs. A focused FM-only test further highlights how modulation type and network architecture influence classifier robustness. AIMC-Spec establishes a reproducible baseline and provides a foundation for future research and standardization in the AIMC domain.

Problem

• Automatic Intrapulse Modulation Classification
• Radar Signal Analysis
• Variable Noise Conditions
• Standardized Dataset
• Signal-to-Noise Ratio

Innovation

• AIMC-Spec
• intrapulse modulation classification
• spectrogram-based dataset
• variable SNR
• deep learning benchmark