Large Wireless Localization Model (LWLM): A Foundation Model for Positioning in 6G Networks
IEEE Transactions on Wireless Communications · 2026 · 25 · 21449-21464
DOI: 10.1109/twc.2026.3718678 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Wireless Communications, 2026. Verfasst von Guangjin Pan und 5 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Chalmers AI Research Center Consortium, EU’s Horizon Europe research and innovation Program, Swedish Foundation for Strategic Research.
Abstract (Originalsprache)
Accurate and robust localization is a critical enabler for emerging 5G and 6G applications, including autonomous driving, extended reality (XR), and smart manufacturing. While data-driven approaches have shown promise, most existing models require large amounts of labeled data and struggle to generalize across deployment scenarios and wireless configurations. To address these limitations, we propose a foundation-model-based solution tailored for wireless localization. We first analyze how different self-supervised learning (SSL) tasks acquire general-purpose and task-specific semantic features based on information bottleneck (IB) theory. Building on this foundation, we design a pretraining methodology for the proposed Large Wireless Localization Model (LWLM). Specifically, we propose an SSL framework that jointly optimizes three complementary objectives: 1) spatial-frequency masked channel modeling (SF-MCM), 2) domain-transformation invariance (DTI), and 3) position-invariant contrastive learning (PICL). These objectives jointly capture the underlying semantics of wireless channel from multiple perspectives. We further design lightweight decoders for key downstream tasks, including time-of-arrival (ToA) estimation, angle-of-arrival (AoA) estimation, single base station (BS) localization, and multiple BS localization. Comprehensive experimental results confirm that LWLM consistently surpasses both model-based and supervised learning baselines across all localization tasks. In particular, LWLM achieves 26.0%–87.5% improvement over transformer models without pretraining, and exhibits strong generalization under label-limited fine-tuning, unseen BS configurations and cross-dataset evaluations, confirming its potential as a foundation model for wireless localization.
Förderung
- Chalmers AI Research Center Consortium
- EU’s Horizon Europe research and innovation Program
- Swedish Foundation for Strategic Research
- Chalmers Areas of Advance in ICT and Transport
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- Vetenskapsrådet
Zitieren
APA
Pan, G., Huang, K., Chen, H., Zhang, S., Hager, C., & Wymeersch, H. (2026). Large Wireless Localization Model (LWLM): A Foundation Model for Positioning in 6G Networks. IEEE Transactions on Wireless Communications, 25, 21449-21464. https://doi.org/10.1109/twc.2026.3718678
Vancouver
Pan G, Huang K, Chen H, Zhang S, Hager C, Wymeersch H. Large Wireless Localization Model (LWLM): A Foundation Model for Positioning in 6G Networks. IEEE Transactions on Wireless Communications. 2026;25:21449-21464. doi:10.1109/twc.2026.3718678
BibTeX
@article{Pan2026,
author = {Pan, Guangjin and Huang, Kaixuan and Chen, Hui and Zhang, Shunqing and Hager, Christian and Wymeersch, Henk},
title = {Large Wireless Localization Model (LWLM): A Foundation Model for Positioning in 6G Networks},
journal = {IEEE Transactions on Wireless Communications},
year = {2026},
volume = {25},
pages = {21449--21464},
doi = {10.1109/twc.2026.3718678}
}
EUVIMED