Improved sensor selection method during movement for breathing rate estimation with unobtrusive pressure sensor arrays
Use of pressure sensor arrays as an unobtrusive way of monitoring physiological characteristics of human beings is a growing field of research. In such pressure sensor arrays, monitoring respiratory signal during movement is a challenge that researchers and engineers are currently faced with. This paper presents an improved method to reliably find breathing rate during movement using unobtrusive bed-based pressure-sensor array. We use spectral flatness ratio and signal variance to select the most powerful sensors for which breathing dominates the signals. We also apply movement detection prior to breathing rate estimation based on a recently developed movement detection algorithm in order to minimize movement effects. The proposed method was applied to nocturnal data collected from a male subject and a female subject, and performance is analyzed. Our results show that this scheme can lead to a higher reliability of estimate even when more than 50% of data are corrupted with movement.
|, , ,|
|2018 IEEE Sensors Applications Symposium, SAS 2018|
|Organisation||Department of Systems and Computer Engineering|
Gilakjani, S.S. (S. Soleimani), Azimi, H. (H.), Bouchard, M. (M.), Goubran, R, & Knoefel, F. (2018). Improved sensor selection method during movement for breathing rate estimation with unobtrusive pressure sensor arrays. In 2018 IEEE Sensors Applications Symposium, SAS 2018 - Proceedings (pp. 1–6). doi:10.1109/SAS.2018.8336769