Wireless fidelity (Wi-Fi) sensing for human gesture detection & recognition

• Main Author: Toh, Douglas Zheng Xun
• Other Authors: Luo Jun
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Wi-Fi sensing; Neural network
• Online Access: https://hdl.handle.net/10356/180993
• Institution: Nanyang Technological University

This research examines the application of Wireless Fidelity (Wi-Fi) Sensing for the detection and identification of human gestures and positions. With the extracted Channel State Information (CSI), the data is processed through a neural network, which is utilized to capture and analyse gestures or positions by learning it's spatial hierarchies. This is all done so by the collection of data via two laptops equipped with Intel AX210 Wi-Fi Network Interface Controllers (NICs) and PicoScenes for CSI extraction. The main neural network used in this research will be the Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). Both networks will be evaluated based on their detection accuracy. This report entails the data collection, experimentation process and results regarding the topic.