Interference Reduction using Bispectrum Estimation in Non-Contact Heart Rate Measurement by Doppler Radar

Non‐contact monitoring of human heartbeat signals using mm‐wave frequency‐modulated continuous‐wave radar under low signal‐to‐noise ratio conditions

“…It is clear from Figure 10 that the proposed method outperforms the methods proposed in Ref. [15,32] under moderate noise with higher accuracy and lower error. It can be seen that the estimation accuracy obtained by the proposed method is higher than 70% when the value of SNR is greater than −10 dB, which shows a greater improvement compared with the other methods, as shown in Figure 10a.…”

“…In it is clear from Table 3 that the proposed coarse-to-fine estimation method consistently outperforms the method in Refs. [15,32] in terms of both the estimation accuracies and estimation errors.…”

“…When the window size is F I G U R E 9 Estimation performance of the proposed method, the frequency-based method in Ref. [32] and the MLE-based method in Ref. [15] versus different length of the heartbeat waveform: (a) the estimation accuracy and (b) the estimation error.…”

“…In order to demonstrate the effectiveness and advantages of the proposed method, in this sub‐section, the performance of the proposed method on the data collected from the participant is shown and a comparison between the methods in Ref. [32] and Ref. [15] is performed.…”

“…The method in Ref. [32] is a typical frequency‐domain‐based algorithm, which estimates the HR by searching the heartbeat peak in the 2D FFT spectrum, and the method proposed in Ref. [15] realises the estimation of HR by the MLE algorithm.…”