Intelligent Real-time Human Activity Recognition Using Wi-Fi Signals - Université de technologie de Troyes Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Intelligent Real-time Human Activity Recognition Using Wi-Fi Signals

Résumé

Human Activity Recognition (HAR) is an important task in many residential health care applications, such as real-time fall detection or elderly health monitoring. This paper proposes a novel HAR solution based on the fine-grained Channel State Information (CSI) of Wi-Fi signals. The proposed approach uses the Deep Learning techniques with the Wi-Fi communication technology to propose an Intelligent Real-time Human Activity Recognition Methodology (IR-HAR). A 2D-layer deep Convolutional Neural Network (CNN) model is thus designed for the classification of raw CSI data generated during human activities. The CNN model is coupled to Time Series Data Augmentation to overcome the small-sized dataset problem. Applied to unseen data, IR-HAR achieves activity recognition for a single user with high accuracy of more than 90%, outperforming the state-of-the-art approaches.
Fichier non déposé

Dates et versions

hal-04444800 , version 1 (07-02-2024)

Identifiants

Citer

Hadi El Zein, Farah Mourad-Chehade, Hassan Amoud. Intelligent Real-time Human Activity Recognition Using Wi-Fi Signals. 2023 International Conference on Control, Automation and Diagnosis (ICCAD), May 2023, Rome, Italy. pp.1-5, ⟨10.1109/ICCAD57653.2023.10152429⟩. ⟨hal-04444800⟩
4 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More