RF-URL 2.0: A General Unsupervised Representation Learning Method for RF Sensing
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2025-10 · 47 (10) · 8889-8906
DOI: 10.1109/tpami.2025.3587718 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025-10. Verfasst von Ruiyuan Song und 7 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 57 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities, Postdoctoral Fellowship Program of China Postdoctoral Science Foundation.
Abstract (Originalsprache)
The major challenge in learning-based RF sensing is acquiring high-quality large-scale annotated datasets. Unlike visual datasets, RF signals are inherently non-intuitive and non-interpretable, making their annotation both time-consuming and labor-intensive. To address this challenge, we propose RF-URL 2.0, a novel unsupervised representation learning (URL) framework for RF sensing, which enables pre-training on easily collected, large-scale unannotated RF datasets to make downstream tasks solve easier. Existing URL techniques, such as contrastive learning, are primarily designed for natural images and are prone to learn shortcuts rather than meaningful information when applied to RF signals. RF-URL 2.0 is the first framework to overcome these limitations by constructing positive and negative pairs through well-established RF signal processing algorithms. Besides, it introduces a novel signal-model-driven augmentation technique, which augments signal representations by identifying and perturbing physically meaningful parameters of signal processing models. Moreover, the RF-URL 2.0 is carefully designed to take into account the heterogeneity characteristics of different RF signal processing representations. We show the universality of RF-URL 2.0 in three typical RF sensing tasks using two general RF devices (WiFi and radar), including human gesture recognition, 3D pose estimation, and silhouette generation. Extensive experiments on the HIBER and WiDAR 3.0 datasets demonstrate that RF-URL 2.0 takes a significant step toward learning-based solutions for RF sensing.
Förderung
- National Natural Science Foundation of China
- Fundamental Research Funds for the Central Universities
- Postdoctoral Fellowship Program of China Postdoctoral Science Foundation
Zitieren
APA
Song, R., Zhang, D., Wu, Z., Yu, C., Xie, C., Yang, S., Hu, Y., & Chen, Y. (2025). RF-URL 2.0: A General Unsupervised Representation Learning Method for RF Sensing. IEEE Transactions on Pattern Analysis and Machine Intelligence, 47(10), 8889-8906. https://doi.org/10.1109/tpami.2025.3587718
Vancouver
Song R, Zhang D, Wu Z, Yu C, Xie C, Yang S, et al. RF-URL 2.0: A General Unsupervised Representation Learning Method for RF Sensing. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2025;47(10):8889-8906. doi:10.1109/tpami.2025.3587718
BibTeX
@article{Song2025,
author = {Song, Ruiyuan and Zhang, Dongheng and Wu, Zhi and Yu, Cong and Xie, Chunyang and Yang, Shuai and Hu, Yang and Chen, Yan},
title = {RF-URL 2.0: A General Unsupervised Representation Learning Method for RF Sensing},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2025},
volume = {47},
number = {10},
pages = {8889--8906},
doi = {10.1109/tpami.2025.3587718}
}
EUVIMED