analiza aiqbench
This study introduces a semi-supervised learning framework using the Mean Teacher architecture to mitigate data collection bottlenecks and improve generalization in dynamic indoor Wi-Fi environments. By leveraging both labeled and unlabeled data for continuous online retraining, the approach achieves significant error reductions compared to conventional supervised methods without incurring excessive computational costs. The technique offers a practical balance between model stability and scalability for large-scale deployment in fluctuating network conditions.
Mean Teacher based SSL Framework for Indoor Localization Using Wi-Fi RSSI Fingerprinting
arXiv:2407.13303v2 Announce Type: replace Abstract: Conventional large-scale indoor localization based on Wi-Fi RSSI fingerprinting faces issues of time-consuming and labor-intensive labeled data collection, limited generalization of a model trained under a supervised learning (SL) framework due to its inability to leverage unlabeled data, and model performance degradation in dynamic scenarios with environmental variations. To address those challenging issues, we propose a comprehensive semi-supervised learning (SSL) framework for a deep neural network (DNN) localization model based on the Mean Teacher, which incorporates access point selection, model pre-training/cloning, and batch-level noise injection. The proposed SSL framework can not only efficiently use hybrid labeled/unlabeled databases for static training of a model during the offline phase, but also exploit unlabeled fingerprints from users of the indoor localization system deployed in the field for continuous retraining of the model during the online phase. We base the proposed SSL framework on the Mean Teacher because it can generate more stable target labels through an exponential moving average of model weights without incurring the high computational complexity of the Pi-Model and with better scalability for online learning than Temporal Ensembling, making it an optimal choice that strikes the right balance between performance and computational complexity in large-scale indoor localization. With the UJIIndoorLoc database, the proposed SSL framework reduces the mean 3D errors of the CNNLoc and SIMO-DNN models by 7.403% and 7.748%, respectively, compared with those under the conventional SL framework; with the XJTLU dynamic database, the maximum reduction in mean 2D error reaches up to 49.227% under a dynamic training scenario, demonstrating the substantial performance improvement achieved by the proposed SSL framework.
Abstract:Conventional large-scale indoor localization based on Wi-Fi RSSI fingerprinting faces issues of time-consuming and labor-intensive labeled data collection, limited generalization of a model trained under a supervised learning (SL) framework due to its inability to leverage unlabeled data, and model performance degradation in dynamic scenarios with environmental variations. To address those challenging issues, we propose a comprehensive semi-supervised learning (SSL) framework for a deep neural network (DNN) localization model based on the Mean Teacher, which incorporates access point selection, model pre-training/cloning, and batch-level noise injection. The proposed SSL framework can not only efficiently use hybrid labeled/unlabeled databases for static training of a model during the offline phase, but also exploit unlabeled fingerprints from users of the indoor localization system deployed in the field for continuous retraining of the model during the online phase. We base the proposed SSL framework on the Mean Teacher because it can generate more stable target labels through an exponential moving average of model weights without incurring the high computational complexity of the Pi-Model and with better scalability for online learning than Temporal Ensembling, making it an optimal choice that strikes the right balance between performance and computational complexity in large-scale indoor localization. With the UJIIndoorLoc database, the proposed SSL framework reduces the mean 3D errors of the CNNLoc and SIMO-DNN models by 7.403% and 7.748%, respectively, compared with those under the conventional SL framework; with the XJTLU dynamic database, the maximum reduction in mean 2D error reaches up to 49.227% under a dynamic training scenario, demonstrating the substantial performance improvement achieved by the proposed SSL framework.
Machine Learning (cs.LG)
arXiv:2407.13303 [cs.LG]
(or arXiv:2407.13303v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2407.13303 Focus to learn more arXiv-issued DOI via DataCite
https://doi.org/10.1016/j.asoc.2026.115711 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 363 cuvinte
Subiecte conexe
Large Scale Entanglement Structure Detection in 100-Qubit Systems via Local Joint Measurements
arXiv · 21 aug. 2026
Researchers Detect Entanglement Patterns in 100-Qubit Systems
Quantum Zeitgeist · 4 sept. 2026
Interpretable Physics Informed WiFi Indoor Localization: Learning an Effective Access Point Geometry and Using It to Prune
arXiv · 30 sept. 2026
STRIDE: State-Transition Representation via Increment Dynamics and Evolution
arXiv · 30 sept. 2026