Deakin University
Browse

File(s) under permanent embargo

WiseFi: activity localization and recognition on commodity off-the-shelf WiFi devices

Version 2 2024-06-06, 05:43
Version 1 2023-10-25, 23:23
conference contribution
posted on 2024-06-06, 05:43 authored by D Zhu, N Pang, Gang LiGang Li, S Liu
Most recently, activity localization and recognition has increasingly attracted significant attentions due to its broad range of applications to support smart devices. Pioneer systems based on WiFi signals usually require six to eight antennas to localize the activity while the commodity WiFi infrastructure does not meet this requirement. In addition, they also require the priori learning of wireless signals to recognize a pre-defined set of activities. In this paper, we present WiseFi, an activity localization and recognition system by leveraging fine-grained physical layer information on commodity off-the-shelf (COTS) WiFi devices. WiseFi harnesses the amplitude and the phase of Channel State Information (CSI), and the Angle-of-arrival (AOA) of blocked signals to localize and recognize human activity. The intuition behind WiseFi is that whenever the target occludes the incoming wireless signals, the power of AOA will drop in the same direction. Experimental results indicate that WiseFi can achieve comparable performance in activity localization and recognition on COTS WiFi devices.

History

Pagination

562-569

Location

Sydney, N.S.W.

Start date

2016-12-12

End date

2016-12-14

ISBN-13

9781509042968

Language

eng

Publication classification

E Conference publication, E1 Full written paper - refereed

Copyright notice

2016, IEEE

Editor/Contributor(s)

O'Conner L

Title of proceedings

HPCC/Smart City/DSS 2016 : Proceedings of the 18th IEEE International Conference on High Performance Computing and Communications, 14th IEEE International Conference on Smart City and 2nd IEEE International Conference on Data Science and Systems

Event

IEEE Computer Society. Conference (18th, 14th, 2nd : 2016 : Sydney, N.S.W.)

Publisher

Institute of Electrical and Electronics Engineers

Place of publication

Piscataway, N.J.

Series

IEEE Computer Society Conference