WiWho: WiFi-Based Person Identification in Smart Spaces

Zeng, Yunze; Pathak, Parth H.; Mohapatra, Prasant

doi:10.1109/ipsn.2016.7460727

Cited by 325 publications

(194 citation statements)

References 20 publications

Supporting

Mentioning

190

Contrasting

Unclassified

Order By: Relevance

“…In Wang et al WCI used to detect the human activity of falling (WiFall) and accomplished 87% accuracy. In Zeng et al they build a system (WiWho) that uses Wi‐Fi WCI signals to recognize gait of a person and for the group of two or six people, accomplished 92% to 80% average accuracy accordingly. Compared with WiWho and WiFall, we use advance algorithms to recognize gaits at a long distance to the line‐of‐sight (LOS), which help to get better accuracy.…”

Section: Related Workmentioning

confidence: 99%

“…Compared with the other gait‐based human recognition systems, wireless sensing system is better technology in our everyday life, wireless technology has been extensively used, and because of its demand and low‐cost deployment, researchers love to use this technology in different kinds of recognition work. With the capability of wireless channel information (WCI), most wireless devices sustain machine‐free recognition . The case of a communication link from the transmitter source to the receiver source produced the wireless channel information (WCI).…”

Section: Preliminariesmentioning

confidence: 99%

See 1 more Smart Citation

Gait signals classification and comparison

Barua

Yang

Ren

et al. 2019

Int J Numerical Modelling

View full text Add to dashboard Cite

Use of wireless signal technology in sensing of human gait activity is a satisfactory example of device‐free sensing and effective in medical science to detect human motion–related diseases. Some prior research showed some potential detecting process of human walking gait from wireless channel information (WCI) using wireless signals. In this paper, we present comparison of three popular features reduction methods such as principal component analysis (PCA), kernel principal component analysis (KPCA), and linear discriminant analysis (LDA) using three classifications methods, support vector machine (SVM), k‐nearest neighbor (k‐NN), and decision tree (DT) in an absolutely equivalent situation for identifying walking gait signals. The analysis was carried out on the WCI‐based dataset where dataset was divided into four classes (normal gait, small gait, fast gait, and turn gait). Using dataset with the combination of methods (features reduction and classification), experimental results shows that all the combinations of PCA, KPCA, and LDA with three classifications achieve an average accuracy of gait identification is accordingly 86%, 79%, and 95%.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Preliminariesmentioning

confidence: 99%

Gait signals classification and comparison

Barua

Yang

Ren

et al. 2019

Int J Numerical Modelling

View full text Add to dashboard Cite

show abstract

“…Han et al [3] proposed to exploit the special diversity of CSI to recognize human fall. Zeng et al [5] presented WiWho that exploits CSI-based gait to identify a person. Wang et al [17] presented locationoriented activity recognition system to recognize bathing and washing dishes.…”

Section: B Wireless Based Activity Detectionmentioning

confidence: 99%

“…Received Signal Strength Indicator (RSSI) based techniques suffer from the performance degradation in indoor environments due to interference or multipath fading. More recently, researchers have resorted to Channel State Information (CSI), which reflects channel frequency response in IEEE 802.11 a/g/n, to recognize some simple activities such as falling down [3], smoking [4], human identification [5]. However, the performance of existing CSI solutions is usually with too high false alarm rates or too low detection recall rate [6].…”

Section: Introductionmentioning

confidence: 99%

NotiFi: A ubiquitous WiFi-based abnormal activity detection system

Zhu

Pang

et al. 2017

2017 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Abstract-Abnormal activity detection has increasingly attracted significant research attention due to its potential applications in numerous scenarios, such as patient monitoring, health care of children and elderly, military surveillance, etc. Pioneer systems usually rely on computer vision or wearable sensors which pose unacceptable privacy risks, or wireless signals which require the priori learning of wireless signals to recognize a set of predefined activities. In this paper, we take the first attempt to achieve non-invasive abnormal activity detection with only commodity off-the-shelf (COTS) WiFi devices, namely NotiFi, that can accurately detect the abnormal activities. The intuition of NotiFi is that whenever the human body occludes the wireless signal transmitting from the access point to the receiver, the phase and the amplitude information of Channel State Information (CSI) will experience a sensitive variation. By creating a multiple hierarchical Dirichlet processes, NotiFi automatically learn the number of human body activity categories for abnormal detection. Extensive experiments in typical realworld environments indicate that NotiFi can achieve satisfactory performance in abnormal activity detection.

show abstract

“…These device-oriented systems offer a means to locate individuals and then occupancy counting is a byproduct, but this approach does pose a burden on the occupant of carrying a specific device. Other designs (e.g, Woyach et al (2006), Xu et al (2013), Zeng et al (2016)) free the occupant from this burden, because the system deduces occupancy by observing how a person's body influences radio wave propagation between the system's radio transmitters and receivers. In practice, this latter method requires a meticulous survey within the building of how an object at a given location changes radio wave propagation.…”

Section: Related Workmentioning

confidence: 99%

A Framework for Occupancy Tracking in a Building via Structural Dynamics Sensing of Footstep Vibrations

Poston

Buehrer

Tarazaga

2017

Front. Built Environ.

View full text Add to dashboard Cite

Counting the number of occupants in building areas over time-occupancy tracking-provides valuable information for responding to emergencies, optimizing thermal conditions or managing personnel. This capability is distinct from tracking individual building occupants as they move within a building, has lower complexity than conventional tracking algorithms require, and avoids privacy concerns that tracking individuals may pose. The approach proposed here is a novel combination of data analytics applied to measurements from a building's structural dynamics sensors (e.g., accelerometers or geophones). Specifically, measurements of footstep-generated structural waves provide evidence of occupancy in a building area. These footstep vibrations can be distinguished from other vibrations, and, once identified, the footsteps can be located. These locations, in turn, form the starting point of estimating occupancy in an area. In order to provide a meaningful occupancy count, however, it is first necessary to associate discrete footsteps with individuals. The proposed framework incorporates a tractable algorithm for this association task. The proposed algorithms operate online, updating occupancy count over time as new footsteps are detected. Experiments with measurements from a public building illustrate the operation of the proposed framework. This approach offers an advantage over others based on conventional technologies by avoiding the cost of a separate sensor system devoted to occupancy tracking.

show abstract

WiWho: WiFi-Based Person Identification in Smart Spaces

Cited by 325 publications

References 20 publications

Gait signals classification and comparison

Gait signals classification and comparison

NotiFi: A ubiquitous WiFi-based abnormal activity detection system

A Framework for Occupancy Tracking in a Building via Structural Dynamics Sensing of Footstep Vibrations

Contact Info

Product

Resources

About