Opportunistic physical activity monitoring via passive WiFi radar

Li, Wenda; Tan, Bo; Piechocki, Robert J.; Craddock, Ian J

doi:10.1109/healthcom.2016.7749458

Cited by 17 publications

(21 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our work in [17] demonstrates the capability of using narrow band energy harvesting signal for respiration detection. The work in [18] shows the activity recognition and activity level indication capability of WiFi data signal and WiFi beacon signal. The work in [19] explores the unsupervised learning approach for activity recognition.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Passive Radar for Opportunistic Monitoring in E-Health Applications

Tan

Piechocki

2018

IEEE J. Transl. Eng. Health Med.

Self Cite

View full text Add to dashboard Cite

This paper proposes a passive Doppler radar as a non-contact sensing method to capture human body movements, recognize respiration, and physical activities in e-Health applications. The system uses existing in-home wireless signal as the source to interpret human activity. This paper shows that passive radar is a novel solution for multiple healthcare applications which complements traditional smart home sensor systems. An innovative two-stage signal processing framework is outlined to enable the multi-purpose monitoring function. The first stage is to obtain premier Doppler information by using the high speed passive radar signal processing. The second stage is the functional signal processing including micro Doppler extraction for breathing detection and support vector machine classifier for physical activity recognition. The experimental results show that the proposed system provides adequate performance for both purposes, and prove that non-contact passive Doppler radar is a complementary technology to meet the challenges of future healthcare applications.

show abstract

Section: Introductionmentioning

confidence: 99%

“…The work in [19] explores the unsupervised learning approach for activity recognition. However, above systems [17] – [19] are all designed for a specific purpose. While in this paper, we proposed a cognitive mechanism to allow the system can work adaptively for different purposes in the residential house.…”

Section: Introductionmentioning

confidence: 99%

Passive Radar for Opportunistic Monitoring in E-Health Applications

Tan

Piechocki

2018

IEEE J. Transl. Eng. Health Med.

Self Cite

View full text Add to dashboard Cite

show abstract

“…This approach requires the radar operator to have user access to the network, making the system unusable when the AP belongs to a third-party, or an adversary. The WiFi beacon signal has been examined as a viable alternative [6] and [13] as it is constantly transmitted by WiFi APs, although it has a much lower bandwidth. Milani et al [19] concluded that the default beacon signal interval time of 100ms in commodity WiFi APs is not sufficient for WiFi sensing as the Doppler record is illegible and corrupted by noise.…”

Section: Wifi Signal Generationmentioning

confidence: 99%

“…When the WiFi AP is idle, it continually transmits a low-bandwidth beacon signal to broadcast its presence. Some recent studies such as [6] and [13] have focused on using passive radar to exploit the WiFi beacon signal but these studies have all increased the repetition rate of the WiFi beacon bursts from every 100 ms (the default setting) to 20 ms (the maximum allowable periodicity on most WiFi APs). These adjustments to make the system usable would again require a level of cooperation with the WiFi AP (in this case password authentication) and therefore cannot be relied on for many real-world deployments.…”

Section: Introductionmentioning

confidence: 99%

Passive Activity Classification Using Just WiFi Probe Response Signals

Shi

Chetty

Julier

2019

2019 IEEE Radar Conference (RadarConf)

View full text Add to dashboard Cite

Passive WiFi radar shows significant promise for a wide range of applications in both security and healthcare owing to its detection, tracking and recognition capabilities. However, studies examining micro-Doppler classification using passive WiFi radar have relied on manually stimulating WiFi access points to increase the bandwidths and duty-cycles of transmissions; either through file-downloads to generate high data-rate signals, or increasing the repetition frequency of the WiFi beacon signal from its default setting. In real-world scenarios, both these approaches would require user access to the WiFi network or WiFi access point through password authentication, and therefore involve a level of cooperation which cannot always be relied upon e.g. in law-enforcement applications. In this research, we investigate WiFi activity classification using just WiFi probe response signals which can be generated using a low-cost off-the-shelf secondary device (Raspberry Pi) eliminating the requirement to actually connect to the WiFi network. This removes the need to have continuous data traffic in the network or to modify the firmware configuration to manipulate the beacon signal interval, making the technology deployable in all situations. An activity recognition model based on a convolutional neural network resulted in an overall classification accuracy of 75% when trained from scratch using 300 measured WiFi proberesponse samples across 6 classes. This value is then increased to 82%, with significantly less training when adopting a transfer learning approach: initial training using WiFi data traffic signals, followed by fine-tuning using probe response signals.

show abstract

“…The work in [4] shows a potential in using acoustic detection sensor but this lacks in detection range. Alternatively, radar technique has the ability in long-term monitoring under almost all indoor condition and therefore can be applied without those limitations [5].…”

Section: Introductionmentioning

confidence: 99%