An Improved Non-contact Vital Signs Detection Algorithm

Aiming at the problem of respiratory harmonics affecting heart rate estimation, this paper proposes a differential recursive least squares multiple classification (DR-MUSIC) algorithm, which combines recursive least squares and multiple signal classification algorithms to improve the accuracy of heart rate measurements. Firstly, recursive least squares is used to suppress the interference of respiratory harmonics, then differential operations are used to enhance the weak signals. Finally a multiple signal classification algorithm is applied to obtain highly accurate spectrograms of the respiratory and heartbeat signals, and then, the heartbeat ratio is acquired. The practical measured data proves that the algorithm can efficiently suppress the interference of respiratory harmonic noise, and the error rate between the estimated and the reference heart rate is only 1.85%-2.53% in single target multiple measured, which is better than existing algorithms.