Computing Reviews

Widar2.0: passive human tracking with a single Wi-Fi link
Qian K., Wu C., Zhang Y., Zhang G., Yang Z., Liu Y.  MobiSys 2018 (Proceedings of the 16th Annual International Conference on Mobile Systems, Applications, and Services, Munich, Germany,  Jun 10-15, 2018) 350-361, 2018. Type: Proceedings
Date Reviewed: 06/22/21

This work demonstrates that it is possible to track the location of a person, with sub-meter accuracy, using only one single Wi-Fi link between a pair of commercial off-the-shelf (COTS) Wi-Fi devices (“one access point plus one client”). This is accomplished just by capturing and analyzing the signals reflected off the human body, without specialized hardware.

Unlike other solutions that rely on the analysis of a single parameter of the signal on multiple Wi-Fi links, Widar2.0 advances the state of the art by combining multiple signal parameters (namely time-of-flight, angle-of-arrival, Doppler frequency shift, and attenuation).

An algorithm has been designed for joint signal parameter estimation, based on a unified mathematical model that represents how the parameters of the signal reflected by the target object change when the target moves. A previous cleaning step is proposed to eliminate the phase noise typically found in channel state information (CSI) measurements on commercial Wi-Fi, for all signal parameters. Subsequently, novel algorithms are proposed to derive precise locations and motions from the obtained parameter estimates, which are initially erroneous due to the cluttering of parameters of multiple reflections.

The empirical evaluation of Widar2.0 provides results with an accuracy (average localization error of 0.75m) that surpasses or is comparable to state-of-the-art methods requiring multiple Wi-Fi links. Together with the method’s strengths, the authors also point out its limitations.

It is a quite technical paper, addressed to researchers in the field of wireless sensing, but nevertheless clear and detailed enough to be understandable by nonexperts with a strong mathematical background. I would definitely recommend the paper as a remarkable contribution to the state of the art in location tracking.

Reviewer:  Angelica de Antonio Review #: CR147291

Reproduction in whole or in part without permission is prohibited.   Copyright 2021 ComputingReviews.com™
Terms of Use
| Privacy Policy