End-to-End WiFi CSI Sensing Framework for Accurate and Privacy-Preserving Activity Recognition
Author:
- 18 December 2025
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Abstract
Recognizing human activities in indoor environments is a key component in smart home and surveillance systems. In this study, we introduce an innovative method for activity recognition using Channel State Information (CSI) of WiFi signals. Compared with traditional approaches that rely on wearable devices or cameras, our method utilizes existing WiFi infrastructure, ensuring privacy and minimal user intrusion. By analyzing the changes in CSI induced by human motions, we develop a robust model capable of accurately identifying a range of activities. Extensive experiments demonstrate the effectiveness of our approach in various indoor settings, highlighting its promise for broad implementation in practical scenarios.Keywords
This publication has 17 references indexed in Scilit:
- Towards a low-cost and privacy-preserving indoor activity recognition system using wifi channel state informationMultimedia Tools and Applications, 2025
- CSHA-CSI: Towards Contactless Sensing of Human Activities Utilizing WiFi Channel State Information2024
- A novel approach for simultaneous human activity recognition and pose estimation via skeleton-based leveraging WiFi CSI with YOLOv8 and mediapipe frameworksSignal, Image and Video Processing, 2024
- Leveraging WiFi Channel State Information for Efficient and Reliable Device-Free Human Activity Recognition2023
- Design and Deployment of a Practical Wireless Sensing System for HAR with WiFi CSI in the 5GHz Band2023
- An Intelligent Monitoring Approach Based on WiFi Sensing for Smart Hospital2023
- A comprehensive review of wifi sensing technologies: Tools, challenges and future research directionsAIP Conference Proceedings, 2023
- CSI-MIMO: An efficient Wi-Fi fingerprinting using Channel State Information with MIMOPervasive and Mobile Computing, 2015
- A Wearable Gesture Recognition Device for Detecting Muscular Activities Based on Air-Pressure SensorsIEEE Transactions on Industrial Informatics, 2015
- Camera-Based Fall Detection on Real World Data2012