Paper
7 August 2024 Real-time random body motion elimination based on phase compensation and recurrent neural network prediction
Chao Wang, Aomei Zheng, Enze Liu, Jiachang Guo, Lianglin Qu, Weinxing Zhang, Pengsong Duan, Yangjie Cao
Author Affiliations +
Proceedings Volume 13224, 4th International Conference on Internet of Things and Smart City (IoTSC 2024); 132240A (2024) https://doi.org/10.1117/12.3034991
Event: 4th International Conference on Internet of Things and Smart City, 2024, Hangzhou, China
Abstract
In this paper, we propose a low complexity method to eliminate random object motion in real time. This method is based on the radar system composed of AD8302 and phase shifter. RNN is used to predict the output voltage signal of AD8302, and the phase difference caused by motion is predicted and compensated in real time according to the relationship of phase voltage, so as to obtain vital signs. At the same time, we also modify the phase voltage curve of AD8302 to increase the accuracy of eliminating random body motion. The experiment shows that the method can eliminate the linear motion signal and retain the sinusoidal signal of the loudspeaker in the case of the combination of linear and sinusoidal motion, which greatly validates the feasibility of the method to eliminate human motion and extract vital signs.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Chao Wang, Aomei Zheng, Enze Liu, Jiachang Guo, Lianglin Qu, Weinxing Zhang, Pengsong Duan, and Yangjie Cao "Real-time random body motion elimination based on phase compensation and recurrent neural network prediction", Proc. SPIE 13224, 4th International Conference on Internet of Things and Smart City (IoTSC 2024), 132240A (7 August 2024); https://doi.org/10.1117/12.3034991
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Radar signal processing

Phase shifts

Vital signs

Data modeling

Radar

Phase compensation

Signal processing

Back to Top