Grouping Similar Sensors Based on their Sent Data in a Massive IoT Scenario - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2024

Grouping Similar Sensors Based on their Sent Data in a Massive IoT Scenario

Résumé

The expanding Internet of Things landscape, combined with a significant reduction in the cost of connected devices, has enabled the widespread deployment of sensors. These sensors, often situated in close proximity to each other, frequently produce overlapping data. This paper aims to identify such overlaps in sensor data to effectively cluster similar sensors. The benefits of clustering include enhanced detection of sensor failures and a reduction in data transmissions. We examine a typical scenario where sensors are deployed asynchronously, operate for a limited time within the environment, and emit data that is noisy, irregular, and unsynchronized. To overcome these challenges, we have developed a similarity metric that employs interpolation techniques to manage noisy, irregular, and unsynchronized data. This metric supports a hierarchical clustering algorithm featuring a novel linkage method tailored to the dynamic nature of sensor deployments. The goal is to cluster sensors that monitor the same phenomenon, regardless of their active periods not coinciding. Through simulations, we demonstrate the superiority of our method compared to the state-of-the-art Dynamic Time Warping distance and a hierarchical clustering with complete linkage inspired by related works. Our results establish a mean improvement of $23\%$ from our approach in terms of V-Measure. We provide comprehensive experiments assessing the robustness of our solution under various sensor measurement noise levels and employing different stopping criterion strategies.
Fichier principal
Vignette du fichier
17 04 2024Grouping_Similar_Sensors_Based_on_their_Sent_Data-6.pdf (704.57 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04424455 , version 1 (29-01-2024)
hal-04424455 , version 2 (17-04-2024)

Licence

Paternité

Identifiants

  • HAL Id : hal-04424455 , version 2

Citer

Gwen Maudet, Mireille Batton-Hubert, Patrick Maillé, Laurent Toutain. Grouping Similar Sensors Based on their Sent Data in a Massive IoT Scenario. 2024. ⟨hal-04424455v2⟩
3 Consultations
8 Téléchargements

Partager

Gmail Facebook X LinkedIn More