RF-based Indoor Tracking System

Abstract

1. What challenges are there for Indoor Tracking?

Unavoidable multipath distortions makes the typical DOA/AOA/TOF methods failed.
The popular GPS cannot work well also due the lack of LoS singals to the GPS satellites.

2. How do you solve the Multipath issue?

We treat LoS and NLoS signal as a whole and explore the statistical property of the received signal by analog to the TR principle. We proved that the TR is mathematically equivalent to the the autocorrelation function of the CSI (ACF). Thus, by considering the Raleigh Fading, we further proved that the ACF exhibits like a Focusing Ball around the receiver in spatial domain. This Focusing Ball is then used to develop the Active Tracking System (see paper of our group).

3. Does it work for collocated WiFi deployment?

Yes. By exploiting the statistical theory of electromagnetic (EM) waves, we establish a link between the autocorrelation function (ACF) of the physical layer CSI and the speed of a moving object, which embraces the rich-scattering environment typical for indoors to facilitate highly accurate Passive Speed Estimation (see paper of our group).

4. Have you tested your system in practical scenarios?

The active one is being commercialized to track a moving object in supermarkets while the passive one is being commercialized as a fall-detector and Gait Recognition, which have been invited for demo to Apple, HP, Qualcomm, Verizon, Marvell.

For more information, please refer to our (Patent: EP3695783)

Key words: Massive MIMO, Time Reversal, Multipath, Localization and Tracking.