Accurate Indoor Tracking Made Easy and Scalable

Chenshu Wu, Feng Zhang, Beibei Wang, K. J. Ray Liu

University of Maryland, College Park and Origin Wireless Inc.


Overview


Ideally, an Indoor Positioning System (IPS) should satisfy the below requirements to achieve a ubiquitous "Indoor GPS": low(zero)-cost, scalable, accurate, easy-to-deploy, large coverage. Existing solutions, however, do not easily scale with low cost while maintaining high accuracy. For example, the mainstream triangulation-based systems degrade drastically under NLOS conditions and require extremely precise installation of multiple APs, while fingerprint-based methods suffer from low accuracy in practice. We design EasiTrack, an indoor real-time tracking system that achieves decimeter accuracy using a single unknown commodity WiFi AP under both LOS and NLOS conditions and scales to massive buildings and unlimited number of clients with (almost) zero cost. Despite over three decades of research, enabling such capabilities in one system has been regarded as challenging, and we are unaware of any prior IPS works by which the same would have been achieved.

We believe EasiTrack sets the stage for a ubiquitous IPS solution for world-wide deployment, and we are currently commercializing it with Origin Wireless Inc. So far, we have demonstrated our system in the headquarters of many big tech companies, including Apple, HP, Qualcomm, Marvell, Murata, etc., and at CES 2019.



Demo Video


Demo Traces


Notes:

@A* Company, CA (40m x 50m)

@M* Company, CA (50m x 50 m)

@CES 2019, NV (13m x 23m)



Paper


EasiTrack: Decimeter-Level Indoor Tracking with Graph-based Particle Filtering
Chenshu Wu, Feng Zhang, Beibei Wang, K. J. Ray Liu
IEEE Internet of Things Journal (IOTJ), 2019
[PDF]

RF-based Inertial Measurement
Chenshu Wu, Feng Zhang, Yusen Fan, K. J. Ray Liu
ACM SIGCOMM, 2019
[PDF]