Interpretable physics-based wiFi localization improves accuracy and prunes access points
Interpretable Physics Informed WiFi Indoor Localization: Learning an Effective Access Point Geometry and Using It to Prune
Artificial IntelligenceMachine Learning
Summary
Indoor location tracking with WiFi signals can be very accurate but often works like a ‘‘black box’’—hard to understand or reuse what it learns. The authors created a model that not only finds where a person is but also figures out the relative positions of WiFi access points without needing those points’ real locations. This understanding helps the model ignore less useful access points without losing much accuracy. Their method performed better on several WiFi indoor location datasets, helping select better features and boost performance while making the system easier to interpret.
What this means in practice
- •For indoor navigation developers: Improve WiFi-based indoor positioning systems by using interpreted access point layouts to reduce needed signal sources while maintaining accuracy.
- •For smart building system designers: Use geometry-based access point selection for more efficient sensor deployment and maintenance without extensive site surveys.
Authors
Arshia Eftekhari zadeh, Rezvan Nasiri, Hadi Moradi
Abstract
Deep learning models can achieve high accuracy for indoor localization, but their black-box nature limits interpretability and the reuse of learned information. We propose a hierarchical deep learning framework for WiFi fingerprint-based indoor localization that jointly predicts user location and learns an effective geometry of the surrounding access points (APs). Physics-informed decoders infer this geometry directly from RSSI measurements and labelled user positions, without requiring the true AP coordinates during training. The learned geometry is then used to rank and prune APs. On the UJIIndoorLoc dataset, the proposed chained model achieves a mean 3D localization error of 7.07 m, reducing error by 26% to 36% compared with baseline models. Previously published methods evaluated on the same official split report errors 10.6% to 31.0% higher. Pruning 35% or 50% of the APs causes only a small loss in localization accuracy. The inferred geometry also enables Fisher-information-based AP ranking even when fingerprint databases do not contain surveyed AP coordinates. Experiments on the Tampere/TUT and UTSIndoorLoc datasets show that geometry-guided AP selection performs comparably to selectors built directly from labelled data. These results show that physics-informed interpretability can improve indoor localization while also supporting effective feature selection.