WIBECAM: Device Free Human Activity Recognition Through WiFi Beacon-Enabled Camera Mauro De Sanctis, Ernestina Cianca, Simone Di Domenico, Daniele Provenziani, Giuseppe Bianchi, Marina Ruggieri Dept. of Electronic Engeneering, University of Rome Tor Vergata Rome, Italy mauro.de.sanctis@uniroma2.it, ernestina.cianca@uniroma2.it, simone.di.domenico@uniroma2.it, giuseppe.bianchi@uniroma2.it ABSTRACT This paper presents WIBECAM, a Human Activity Recog- nition system which does not require neither user instrumen- tation, nor specialized infrastructure, nor active operation - it passively leverages Beacon frames periodically emitted by a single off-the-shelf Wi-Fi access point. As many other re- cent proposals, WIBECAM also exploits the different multi- path conditions (and their temporal variations) induced by human activity. In most of the previously proposed sys- tems, the classification is based on the characterization of the signal strength variations, caused by the human activity. WIBECAM’s main distinguishing aspect is that it “watches” the channel in the frequency domain where spectral metrics, calculated on the raw signal samples of the received Beacon frames, are like “snapshots” of the channel taken in a regular and periodical way. The classification process uses properly selected features that measure the changes of consecutive “snapshots”. WIBECAM adapts to any Wi-Fi access point (and may comply even with legacy 802.11b-only ones), as it does not exploit neither OFDM and CSI extracted from the receiver, nor MIMO/multiple antennas. WIBECAM has been built into USRP software radios. Its classification accu- racy has been preliminarily assessed for four different activ- ities in two different environments; the resulting confusion matrices show very promising performance. Categories and Subject Descriptors H.4 [Information Systems Applications]: Miscellaneous Keywords Device-free Human Activity Recognition, WiFi, RF sensing. 1. INTRODUCTION This paper presents WIBECAM, an Activity Recognition (AR) system that requires neither user instrumentation nor an infrastructure of cameras. The WIBECAM system makes Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full cita- tion on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or re- publish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. WPA’15, May 22, 2015, Florence, Italy. Copyright c  2015 ACM 978-1-4503-3498-3/15/05 ...$15.00. DOI: http://dx.doi.org/10.1145/2753497.2753499. use of a single Access Point (AP) and a single WiFi receiver with single antennas, in contrast with other recent AR sys- tems that use multiple receivers and/or multiple APs and/or multiple antennas (e.g. [1], [2]). As several other recently proposed AR systems [1], [3], WIBECAM captures WiFi sig- nals in an environment by looking at multi-path distortions or, more precisely, their time variations which are related to human motion in the environment. Key elements of the proposed AR systems are a) use of the Beacon messages of a WiFi AP; b) extraction of the channel frequency behav- ior from the I/Q components of the received signal; c) a videocamera-like approach for automatic AR to catch the time variations of the channel frequency behavior. Use of beacon signals - Our AR system collects the raw signal samples from the Beacon packets of just a single AP. The advantages of using Beacon packets, instead of data packets [4], are manyfold: 1. No need to transmit or receive user-generated data from the AP (hence no waste of channel resources). 2. In contrast with AR systems that require a continuous transmission, we proved that our system can effectively use short length data packets (the Beacon messages) with long inactive periods in the order of tens of ms. 3. No need to authenticate to the AP. 4. It can be applied to any IEEE802.11x standard, in- cluding Orthogonal Frequency Division Multiplexing (OFDM) standards and non-OFDM standards. As shown in Section II, other works also use Beacon mes- sages, but in a different way [2]. Channel behaviour in the frequency domain - The pro- posed system uses the raw signal samples of the received Beacon packets to calculate the channel behavior in the fre- quency domain, which is strictly related to the multipath propagation and hence, to the specific scatterers present in the sensed environment. We further show that the usage of estimated Power Spectral Density (PSD) is a better ap- proach over Discrete Fourier Transform (DFT). Videocamera-like approach - In a videocamera-based AR system, the videocamera captures images at a constant rate of about 15-30 frames per second (fps) and the automatic AR is based on the analysis of the differences between con- secutive captured images. WIBECAM performs the activity recognition by capturing the frequency behavior of the chan- nel, through specific curves (i.e. PSD), at a constant rate 7