Wi-Fi human gesture recognition for real-time system | |
| Author | Tanund Sakpoonsup |
| Call Number | AIT Thesis no.IOT-24-02 |
| Subject(s) | Human-computer interaction Real-time data processing Wireless communication systems |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Internet of Things (IoT) Systems Engineering |
| Publisher | Asian Institute of Technology |
| Abstract | In recent years, the rise of Internet of Things (IoT) devices and real-time interactive sys tems has led to an increased demand for seamless human-computer interaction methods. Traditional approaches for gesture detection frequently rely on cameras or specialized sensors, which impose constraints in various settings such as low-light conditions or privacy problems. Wi-Fi signals, on the other hand, do not have this constraint, mak ing them a non-intrusive and privacy-preserving option. This thesis explores the de velopment of a Wi-Fi-based human gesture recognition for real-time system, leveraging Channel State Information (CSI) data and Convolutional Neural Network (CNN) mod els. The proposed system utilizes low-cost hardware, including a Raspberry Pi equipped with the Nexmon tool, to capture and process CSI data from Wi-Fi signals.The primary objective of this research is to design, implement, and evaluate a robust gesture recognition framework capable of accurately detecting and classifying human gestures to facilitate intuitive real-time system control. To achieve this, a comprehensive dataset of CSI samples is collected through controlled gesture actions, encompassing a range of common human gestures such as drawing simple symbols. The collected data is preprocessed to extract relevant features, such as amplitude and phase information, which are then employed as input to train a CNN model.The findings of this study underscore the feasibility and effectiveness of utilizing CSI data for human gesture recognition. The proposed system demonstrates promising ac curacy rates across a variety of gestures, making it a potential solution for enhancing the user experience in real-time system. Furthermore, the research contributes to the broader field of human-computer interaction by showcasing the viability of exploiting wireless signals beyond their conventional networking applications. In conclusion, this thesis introduces an innovative method for recognizing human ges tures in real-time system using Wi-Fi. Through the integration of affordable hardware and advanced deep learning techniques, this system creates opportunities for seamless and instinctive interactions for smart homes, healthcare, virtual reality, and human computer interaction. The findings and knowledge obtained from this research lay the foundation for exploring more sophisticated applications of wireless signals across dif ferent domains, fostering a deeper connection between humans and technology. |
| Year | 2024 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Internet of Things (IoT) Systems Engineering |
| Chairperson(s) | Attaphongse Taparugssanagorn |
| Examination Committee(s) | Chaklam Silpasuwanchai;Chantri Polprasert |
| Scholarship Donor(s) | Royal Thai Government Fellowship |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2024 |