Paper
14 February 2022 Wi-Res: a human motion recognition method based on CSI signal
Dongzhi He, Longhang Guo, Kai Niu
Author Affiliations +
Proceedings Volume 12161, 4th International Conference on Informatics Engineering & Information Science (ICIEIS2021); 121610N (2022) https://doi.org/10.1117/12.2626839
Event: 4th International Conference on Informatics Engineering and Information Science, 2021, Tianjin, China
Abstract
The action recognition method based on Wi-Fi is one of the research focuses in recent years. Existing methods have problems with complex noise processing and weak classification ability. To solve the limitation of existing human action recognition methods, this paper proposes a human action recognition solution that is a method based on Wi-Fi signal and the residual network (Wi-Res). First, the method captures human motion information (running, walking, sitting, standing, jumping) using CSI (Channel State Information) signals and constructs motion templates based on time series. Second, this work uses the obtained motion templates as basic input units. Third, the Wi-Res extracts feature through convolutional neural networks and adds residual connections to solve problems, such as gradient disappearance. The method can describe motion on a longer time scale. The human action recognition method based on the residual network receives an accuracy rate of about 97% for human action recognition when tested in indoors.
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Dongzhi He, Longhang Guo, and Kai Niu "Wi-Res: a human motion recognition method based on CSI signal", Proc. SPIE 12161, 4th International Conference on Informatics Engineering & Information Science (ICIEIS2021), 121610N (14 February 2022); https://doi.org/10.1117/12.2626839
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Gait analysis

Feature extraction

Signal attenuation

Antennas

Computing systems

Detection and tracking algorithms

Motion models

Back to Top