Implementation of a Heart Rate Detection Scheme Based on Wavelet Packet
The realization of non-contact radar heart rate detection technology is now a research focus. In the process of estimating heart rate by millimeter wave radar, respiratory and its harmonics, clutter, noise, and other interferences degrade the accuracy of heart rate detection greatly. Many schemes have been proposed for dealing with these issues, and are mostly computation intensive. To realize heart rate detection in real-time, this paper designs an accurate heart rate detection scheme with low computational complexity based on wavelet packet decomposition in a millimeter wave radar from TI. The novel scheme uses a six-layer wavelet packet decomposition and selects nodes whose signal frequency varied from 2 Hz to 4 Hz to take part in the reconstruction of the second harmonic of the heartbeat. Nodes attending reconstruction are weighted according to their square root of energy to further improve the signal-to-noise ratio. In addition, an optimization method is proposed to reduce computation burden by 40 percent so as to realize the proposed scheme in real-time. Programs are designed based on modular design method, and realization of some important modulars are described and explained. The whole scheme is realized in a AWR1642 millimeter wave radar from TI by using C language. Experimental results demonstrate that the proposed scheme for heart rate detection achieves an average absolute error of 2.55 beats per minute (bpm), which is 7.74 bpm better than that of the traditional heart rate detection scheme.
Authors
- Jiayang Xiao (ORCID: https://orcid.org/0000-0002-8007-4995)
- Yousheng Wang (ORCID: https://orcid.org/0000-0003-1560-8695)
- Xiaolin Xu (ORCID: https://orcid.org/0000-0001-8393-2783)
- Qitong Li (ORCID: https://orcid.org/0000-0001-8048-9181)
- Yunxue Liu (ORCID: https://orcid.org/0009-0003-4564-3237)
- Yan Zhang (ORCID: https://orcid.org/0000-0001-7644-7480)
Institutions
- Yantai University (CN)
Publication Details
- Published
- 2024-10-26
- DOI
- https://doi.org/10.1109/cisp-bmei64163.2024.10906257
- Primary Topic
- Advanced Computing and Algorithms
- Type
- article
- Field-Weighted Citation Impact
- 0.00