cs.LGNov 11, 2026

Observations Drift, Structures Remain: Structural Pretraining with Time Alignment for Electromagnetic Signals

Authors: Wenjin Gui, Junyu Shen, Haibo Xu, Yuchuang Sun, Luqing Luo, Wupeng Xie, Qian Huang, Fengxiang Wang, 

Abstract

Electromagnetic signal modeling is increasingly moving toward self-supervised pretraining on heterogeneous I/Q observations. However, an observed waveform is jointly shaped by signal generation, propagation environments, and acquisition processes, which mix reusable structural regularities with observation-specific variations. Existing methods usually pursue unified pretraining with reconstruction objectives on mixed observations. This heterogeneity creates two key mismatches for reconstruction pretraining: (1) appearance mismatch, induced by propagation-dependent waveform variations; and (2) temporal mismatch, induced by acquisition-dependent sampling coordinates. These mismatches make observation-specific appearances and sampling coordinates easier to reconstruct than stable signal structures, inducing observational shortcuts. We identify this failure mode as observation-biased pretraining. To address this, we propose Structural Pretraining with Time Alignment for heterogeneous EM signals. It mitigates temporal mismatch with RoPE-fs, which maps token positions to sampling-rate-normalized physical-time positions, and reduces reliance on appearance shortcuts by replacing waveform reconstruction with masked prediction of discrete structural units. These units are organized into atomic structural units and emergent structural units that capture variable-length compositional patterns beyond fixed patch boundaries. Across diverse EM benchmarks, our method improves both full-supervision and few-shot adaptation over existing EM pretraining baselines, showing that the learned representations better preserve structural consistency when observations drift in waveform appearance and temporal scale.

Figures & tables

Appendix figures & tables11 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. LEMMO: A Large Electromagnetic Model leveraging Pretrained LLMs for Million-Sample Signal Understanding

    Nov 6, 2026Mingkai Li, Jiajun Wu, Yihan Mao +6Multimodal Large Language Models

  2. PReD: An LLM-based Foundation Multimodal Model for Electromagnetic Perception, Recognition, and Decision

    Mar 30, 2026Zehua Han, Jing Xiao, Yiqi Duan +13Wireless Foundation ModelsMultimodal Foundation Models

  3. Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning

    May 1, 2026Hao Zhou, Simon A. Lee, Cyrus Tanade +12Representation LearningCross-Modal Representation Learning