WiFi‐based passive sensing system for human presence and activity event classification

Wenda Li (Lead / Corresponding author), Bo Tan, Robert J. Piechocki

Research output: Contribution to journalArticlepeer-review

22 Citations (Scopus)

Abstract

Detection of human presence and activity event classification are of importance to a variety of context-awareness applications such as e-Healthcare, security, and low impact building. However, existing radio frequency identification tags, wearables, and passive infrared approaches require the user to carry dedicated electronic devices that suffer from problems of low detection accuracy and false alarms. This study proposes a novel system for non-invasive human sensing by analysing the Doppler information contained in the human reflections of WiFi signal. Doppler information is insensitive to stationary objects, thus there is no need for any scenario-specific calibration which makes it ideal for human sensing. We also introduce the time-frequency domain feature vectors of WiFi Doppler information for the support vector machine classifier towards activity event recognition. The proposed methodology is evaluated on a software defined radio system together with the experiment of five different events. The results indicate that the proposed system is sufficient for indoor context awareness, with 95.3% overall accuracy for event classification and 93.3% accuracy for human presence detection, which outperforms the traditional received signal strength approach where accuracy is 69.3% for event classification and 83.3% for human presence detection.
Original languageEnglish
Pages (from-to)276-283
Number of pages7
JournalIET Wireless Sensor Systems
Volume8
Issue number6
DOIs
Publication statusPublished - 1 Dec 2018

Fingerprint

Dive into the research topics of 'WiFi‐based passive sensing system for human presence and activity event classification'. Together they form a unique fingerprint.

Cite this