Development of a device free tracking system for human tracking

• Main Author: Zhan, Lijuan
• Other Authors: CHEAH Chien Chern
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Electrical and electronic engineering::Wireless communication systems
• Online Access: https://hdl.handle.net/10356/141291
• Institution: Nanyang Technological University

Wireless technologies have attracted increasing attentions among these years due to its miniaturization, identity and flexibility, which have greatly promoted the quality and convenience of human life in many aspects. Among the wireless technologies, wireless sensor network, a special kind of point-to-point network, usually does not need high transmission bandwidth, and has low transmission delay and extremely low, which has attracted more and more research and investment attention. With the expansion of the elderly group, the security of the elderly at home is also one of the research hotspots. The home camera monitor used mostly cannot protect the privacy of the elderly well, which is increasingly valued in today's society. Our research aims at developing the wireless network system for human tracking without the need of videos or other devices with human body. In this dissertation, the received signal strength indicator (RSSI) are used to detect human with the wireless sensor network topology. Our human detection method does not require an external sensor and can be detected using the signal strength fluctuation of RSSI. This detection method mainly considers the stability of the RSSI data collected and Kalman filtering is used; then, the deep learning algorithm is used to train the prediction model to predict the specific coordinates of the human body in the monitoring space and verify its accuracy through experiments. Finally, the static human detection model is then applied to real-time detection and tracking, and its practicality is verified through a specific moving path. This dissertation discusses the detection and tracking of human body position based on RSSI figures and has achieved some preliminary results. The study of volatility based on RSSI figures has very good application prospects and is worthy of our further attention and research.