Development of a heart rate detection algorithm using a non-contact doppler radar via signal elimination
S. H. Oh, Seungbong Lee, Sung Min Kim, Jae Hoon Jeong
Abstract
Non-contact heart rate measurement is more convenient than conventional ECGs. So, a number of algorithms have been investigated to measure non-contact heart rate using a Doppler radar. However, elimination of environmental noise is still a challenge. We propose algorithms applying new filtering methods to efficiently detect the Heart rate. The Signal Elimination (SE) facilitates noise acquisition gradually and eliminates it from the previous signal. First, low-frequency noise is acquired from the original signal and removed from it. Second, high-frequency noise is acquired and removed from the previous signal. The Heart rate is then detected via differential polarization and threshold values. Experiments to assess the performance of the Heart rate detection algorithm were conducted on five subjects in a shielded room blocked from external electromagnetic wave. The algorithm has a mean sensitivity of 97.8 %, a positive predictivity of 96.2 %, and an error of 1.6 % compared with the Heart rate count in the reference Biopac. The performance of the proposed algorithm was better than that of the recently studied Heart rate detection algorithm using Doppler radar. Compared with Biopac, the Doppler radar facilitated a robust detection of the heart rate suggesting that the Doppler radar is an effective alternative to contact electrocardiogram in patients with burns or physical disorders.