Algorithm 2: Selection of the fast time indexes that gives the best fit signals

Find the sinusoid fit as shown in Algorithm 1, which gives three parameters i.e., magnitude, frequency and phase shift of the sinusoid that best fits the signal. Find the R square values for the signals at each fast time index. If the fitting frequency of a signal is not in the range of the respiration frequencies i.e., 0.15 Hz to 0.5 Hz, as shown in Figure 4a,b then replace its R square value by zero. If the fitting sinusoid magnitude is less than a certain threshold value as shown in Figure 4c then replace its R square value by zero. Figure 4d shows the fitting for the signal that have the vital signs information as it has relatively higher magnitude and its frequency lies in the breathing frequencies range i.e., 0.15 Hz to 0.5 Hz. Apply the moving averaging filter to the resulting R square values as shown in Figure 5. Extract received signals with R values above threshold value.