Radio map reconstruction with deep neural networks in a weakly labeled learning context with use of heterogeneous side information - Laboratoire d'Informatique de Grenoble Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Radio map reconstruction with deep neural networks in a weakly labeled learning context with use of heterogeneous side information

Résumé

Wireless localization is an essential component in Internet of Things applications. In this paper, we address the generalization problem of RSS map reconstruction for each city (usually, over gateways), relying on location-dependent radio measurements and taking advantage of side knowledge about the local region; like, city plan, terrain height, gateway position. We employ Neural Architecture Search to find an optimized Neural Network model with the best architecture for each of the supposed settings, depending on the amount of such prior side information. We show that using additional side information improves the final accuracy of the Received Signal Strength map reconstruction, especially in sub-areas close to the gateways where larger variations of the average received signal power are usually observed (ultimately, with a prominent beneficial impact onto positioning performance accordingly).
Fichier principal
Vignette du fichier
ASPAI.pdf (1.55 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03823629 , version 1 (21-10-2022)

Identifiants

  • HAL Id : hal-03823629 , version 1

Citer

Aleksandra Malkova, Massih-Reza Amini, Benoît Denis, Christophe Villien. Radio map reconstruction with deep neural networks in a weakly labeled learning context with use of heterogeneous side information. ASPAI 2022 - 4th International Conference on Advances in Signal Processing and Artificial Intelligence, International Frequency Sensor Association (IFSA), Oct 2022, Corfu, Greece. ⟨hal-03823629⟩
66 Consultations
55 Téléchargements

Partager

Gmail Facebook X LinkedIn More