EMind: A Foundation Model for Multi-task Electromagnetic Signals Understanding

2025-08-26

AI Summary

Electromagnetic (EM) signals exhibit strong heterogeneity, high background noise, and complex time-frequency structures, leading to poor generalization and limited cross-task transferability of existing models, compounded by the absence of large-scale, standardized, multi-task benchmark datasets. Method: We introduce the first large-scale, comprehensively annotated EM signal benchmark dataset covering diverse tasks including communications and sensing; propose length-adaptive multi-signal packing and hardware-aware pretraining; and design the first foundational model for EM signal understanding, integrating physics-informed priors with contrastive learning for unified representation learning. Contribution/Results: Experiments demonstrate significant improvements across downstream tasks—including channel estimation, modulation classification, and target detection—with an average performance gain of +8.2%, 40% faster convergence in cross-task transfer, and 35% enhanced robustness—advancing EM intelligence from task-specific to general-purpose capabilities.

Technology Category

• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Deep understanding of electromagnetic signals is fundamental to dynamic spectrum management, intelligent transportation, autonomous driving and unmanned vehicle perception. The field faces challenges because electromagnetic signals differ greatly from text and images, showing high heterogeneity, strong background noise and complex joint time frequency structure, which prevents existing general models from direct use. Electromagnetic communication and sensing tasks are diverse, current methods lack cross task generalization and transfer efficiency, and the scarcity of large high quality datasets blocks the creation of a truly general multitask learning framework. To overcome these issue, we introduce EMind, an electromagnetic signals foundation model that bridges large scale pretraining and the unique nature of this modality. We build the first unified and largest standardized electromagnetic signal dataset covering multiple signal types and tasks. By exploiting the physical properties of electromagnetic signals, we devise a length adaptive multi-signal packing method and a hardware-aware training strategy that enable efficient use and representation learning from heterogeneous multi-source signals. Experiments show that EMind achieves strong performance and broad generalization across many downstream tasks, moving decisively from task specific models to a unified framework for electromagnetic intelligence. The code is available at: https://github.com/GabrielleTse/EMind.

Problem

• Addressing electromagnetic signals' heterogeneity, noise, and complex structure
• Overcoming lack of cross-task generalization in electromagnetic systems
• Solving large-scale dataset scarcity for unified electromagnetic learning

Innovation

• Large-scale pretrained foundation model for electromagnetic signals
• Unified standardized dataset covering multiple signal types
• Length-adaptive packing method and hardware-aware training strategy