GLocFM: A Geometry-Aware Foundation Model for 3D Indoor Wireless Localization

2026-08-10

AI Summary

This work addresses the challenge that existing learning-based wireless localization methods struggle to effectively integrate environmental geometric information, limiting their ability to model non-line-of-sight conditions and generalize across scenes. To overcome this, we propose the first geometry-aware foundational model for WiFi-based localization by jointly modeling WiFi measurements and 3D point cloud geometry. Our approach employs a hierarchical scene encoder to extract propagation-relevant features and leverages multimodal synthetic data generated by Sionna RT to explicitly model direct and single-bounce reflection paths. A learnable scoring function is introduced to match observed and predicted delay-angle spectra, enabling robust maximum-likelihood localization even under unknown time offsets. Evaluated on both synthetic and real-world datasets, our method reduces average 3D localization error by 49.5% and 48.8%, respectively, compared to state-of-the-art approaches, demonstrating its effectiveness and strong cross-scenario generalization.

Abstract

Learning-based wireless localizers often fail to utilize geometric information about the propagation environment, limiting their ability to exploit non-line-of-sight (NLoS) propagation and generalize across scenes. To bridge this gap, we propose GLocFM, a Geometry-aware Localization Foundation Model, which jointly exploits WiFi measurements and scene geometry represented as a 3D point cloud. We formulate localization as a maximum-likelihood (ML) estimation problem, where the goal is to find a transmitter position that maximizes the likelihood of the wireless observations conditioned on the scene geometry. The likelihood of a candidate transmitter position is calculated by a learned scoring function that matches the observed delay--angle-of-arrival (AoA) spectrum against the spectrum predicted for that candidate. A hierarchical scene encoder extracts propagation-relevant features to produce geometric priors for LoS and one-bounce reflection paths. For scenarios with imperfect synchronization, we further introduce a time-of-flight (ToF)-robust GLocFM model to handle unknown ToF offsets. GLocFM is trained on a multi-modal synthetic indoor localization dataset comprising 221 diverse scenes whose associated wireless signals are generated using Sionna RT. On both synthetic and the NeRF$^{2}$ dataset based on real measurements, GLocFM reduces mean 3D localization error relative to one of the state-of-the-art localization baselines by 49.5\% and 48.8\%, respectively. Ablations across different number of receiver, bandwidths, and array sizes further demonstrate the effectiveness and robustness of the proposed framework.

Problem

• wireless localization
• geometry-aware
• non-line-of-sight
• 3D indoor
• scene generalization

Innovation

• Geometry-aware localization
• 3D point cloud
• maximum-likelihood estimation
• delay-AoA spectrum
• ToF-robust model