Our Research on WiFi Sensing

WiFi Sensing and Localization

Exploit the received WiFi signals for sensing and locating objects in the environment. We target to develop an integrated solutions for smart home applications for safety, security, health, and entertainment. We have now developed two real-time demonstrators that can do human behaviour recognition and locating and tracking moving objects.

Demo for Activity recognition.mp4

WiFi sensing for behaviour recognition

This video demonstrates our WiFi behaviour recognition system that can successfully detect several human behaviours including falling down. It uses 802.11n signals collected from three receiver antennas of a COTS WiFi network interface card (one transmitter antenna, 40MHz bandwidth, 5GHz band), without requiring dedicated devices.

Main Publications:

Zhenguo Shi, J. Andrew Zhang, Richard Xu and Qingqing Cheng, “Environment-Robust Device-free Human Activity Recognition with Channel-State-Information Enhancement and One-Shot Learning”, IEEE Trans. On Mobile Computing, 2020.
Zhenguo Shi, Qingqing Cheng, J. Andrew Zhang, Richard Xu, “Environment-Robust WiFi-based Human Activity Recognition using Enhanced CSI and Deep Learning”, IEEE Internet-of-Things Journal, 2022

wifi_tracking_lowReso.mp4

WiFi Localization and Tracking

This video demonstrates our work on localization and tracking using the same setup as above. No fingerprinting technique is used, and the system works by only requiring that the relative position of the transmitter is known to the receiver. The system can be deployed to a new environment without requiring any calibration and training.

Our next step is to extend its capability to multi-objects and through-the-wall tracking.

Zhongqin Wang, J. Andrew Zhang, Ming Xu, and Y. Jay Guo, “Single-Target Real-Time Passive WiFi Tracking”, IEEE Trans. On Mobile Computing, 2021