Privacy-Preserving Non-Wearable Occupancy Monitoring System Exploiting Wi-Fi Imaging for Next-Generation Body Centric Communication

Micromachines

Fawad Ahmed¹  Syed Aziz Shah²  William Buchanan³  Jawad Ahmad³  Gordon Russel³  SyedYaseen Shah⁴  QammerH. Abbasi⁵  Ahsen Tahir⁶ 

[1] Department of Electrical Engineering, HITEC University Taxila, Punjab 47080, Pakistan;School of Computing and Mathematics, Manchester Metropolitan University, Manchester M13 9PL, UK;School of Computing, Edinburgh Napier University, Edinburgh EH10 5DT, UK;School of Computing, Engineering and Built Environment, Glasgow Caledonian University, Glasgow G4 0BA, UK;School of Engineering, University of Glasgow, Glasgow G12 8QQ, UK;University of Engineering and Technology, Lahore, Punjab 39161, Pakistan;

关键词: Wi-Fi;    Privacy;    Occupancy;    Deep Learning;    Encryption;   

DOI  :  10.3390/mi11040379

来源: DOAJ

【 摘 要 】

Nano-scaled structures, wireless sensing, wearable devices, and wireless communications systems are anticipated to support the development of new next-generation technologies in the near future. Exponential rise in future Radio-Frequency (RF) sensing systems have demonstrated its applications in areas such as wearable consumer electronics, remote healthcare monitoring, wireless implants, and smart buildings. In this paper, we propose a novel, non-wearable, device-free, privacy-preserving Wi-Fi imaging-based occupancy detection system for future smart buildings. The proposed system is developed using off-the-shelf non-wearable devices such as Wi-Fi router, network interface card, and an omnidirectional antenna for future body centric communication. The core idea is to detect presence of person along its activities of daily living without deploying a device on person’s body. The Wi-Fi signals received using non-wearable devices are converted into time–frequency scalograms. The occupancy is detected by classifying the scalogram images using an auto-encoder neural network. In addition to occupancy detection, the deep neural network also identifies the activity performed by the occupant. Moreover, a novel encryption algorithm using Chirikov and Intertwining map-based is also proposed to encrypt the scalogram images. This feature enables secure storage of scalogram images in a database for future analysis. The classification accuracy of the proposed scheme is 91.1%.

【 授权许可】

Unknown