Large Wireless Localization Model (LWLM): A Foundation Model for Positioning in 6G Networks

2025-05-15

AI Summary

Current 6G wireless localization faces critical bottlenecks—strong reliance on labeled data and poor generalization across diverse scenarios and base station configurations. Method: We propose the first universal foundation model for 6G localization, featuring a novel three-objective self-supervised pretraining framework that jointly integrates (i) spatial-frequency masked channel modeling, (ii) domain-transformation invariance constraints, and (iii) location-agnostic contrastive learning—guided by information bottleneck theory for multi-task joint pretraining—and a lightweight decoder unifying ToA/AoA estimation with single- and multi-base-station localization. Results: Our model significantly outperforms baselines across all localization tasks, achieving 26.0%–87.5% improvement over non-pretrained Transformers. It demonstrates strong generalization under low-label fine-tuning and unseen base station configurations, establishing a new paradigm for semantic representation of wireless channels.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Intelligent Robots: Localization, Mapping, and Navigation
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

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: (i) spatial-frequency masked channel modeling (SF-MCM), (ii) domain-transformation invariance (DTI), and (iii) 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 and unseen BS configurations, confirming its potential as a foundation model for wireless localization.

Problem

• Addresses lack of generalization in wireless localization models
• Reduces dependency on labeled data for localization tasks
• Improves accuracy across diverse deployment scenarios

Innovation

• Self-supervised learning framework for wireless localization
• Joint optimization of three complementary SSL objectives
• Lightweight decoders for key downstream localization tasks