Radar vital signal tracking using a Kalman filter

1. A method comprising:

receiving reflected radar signals with a millimeter-wave radar;

generating a displacement signal indicative of a displacement of a target based on the reflected radar signals;

filtering the displacement signal using a bandpass filter to generate a filtered displacement signal;

determining a first rate indicative of a heartbeat rate of the target based on the filtered displacement signal;

tracking a second rate indicative of the heartbeat rate of the target with a track using a Kalman filter;

determining a gating region using a gating function;

updating the track when the first rate is inside the gating region; and

updating a setting of the bandpass filter to narrow a pass-band of the bandpass filter based on the updated track.

2. The method of claim 1 , wherein the displacement signal is a time-domain signal.

3. The method of claim 1 , wherein the gating function is an ellipsoidal gating function.

4. The method of claim 1 , further comprising:

incrementing a counter when the first rate is outside the gating region; and

increasing the pass-band of the bandpass filter when the counter reaches a count of M, wherein M is a positive integer greater than 1.

5. The method of claim 4 , wherein increasing the pass-band of the bandpass filter comprises increasing the pass-band based on the track.

6. The method of claim 1 , wherein updating the setting of the bandpass filter based on the updated track comprises adjusting a lower cutoff frequency of the bandpass filter or a higher cutoff frequency of the bandpass filter based on an uncertainty value associated with the updated track.

7. The method of claim 1 , wherein updating the setting of the bandpass filter based on the updated track comprises adjusting a center frequency of the bandpass filter.

8. The method of claim 1 , wherein the bandpass filter has an initial pass-band from 0.7 Hz to 3 Hz.

9. The method of claim 1 , wherein generating the displacement signal comprises:

generating range data based on the reflected radar signals;

performing target detection based on the range data to detect the target; and

generating the displacement signal based on in-phase (I) and quadrature (Q) signals associated with the detected target.

10. The method of claim 9 , wherein the range data corresponds to a range-Doppler image or a range cross-range image.

11. The method of claim 9 , wherein performing target detection comprises determining a target range bin, and wherein generating the displacement signal comprises generating the displacement signal when a standard deviation in the target range bin is below a predetermined threshold.

12. The method of claim 1 , wherein estimating the first rate comprises counting peaks in the filtered displacement signal.

13. The method of claim 1 , wherein the target is a human target.

14. The method of claim 1 , further comprising transmitting radar signals with the millimeter-wave radar, wherein the transmitted radar signals comprise linear chirps.

15. A device comprising:

a millimeter-wave radar configured to transmit chirps and receive reflected chirps; and

a processor configured to:

generate a displacement signal indicative of a displacement of a target based on the reflected chirps,

filter the displacement signal using a bandpass filter to generate a filtered displacement signal,

determine a first rate indicative of a heartbeat rate of the target based on the filtered displacement signal,

track a second rate indicative of the heartbeat rate of the target with a track using a Kalman filter,

determine a gating region using a gating function,

update the track when the first rate is inside the gating region, and

update a setting of the bandpass filter to narrow a pass-band of the bandpass filter based on the updated track.

16. The device of claim 15 , wherein the millimeter-wave radar comprises the processor.

17. The device of claim 15 , wherein updating the setting of the bandpass filter based on the updated track comprises adjusting a lower cutoff frequency of the bandpass filter or a higher cutoff frequency of the bandpass filter based on an uncertainty value associated with the updated track.

18. A method for tracking a vital rate indicative of a heartbeat rate of a human target using a track of a Kalman filter, the method comprising:

receiving reflected radar signals;

generating a displacement signal indicative of a displacement of a target based on the reflected radar signals;

filtering the displacement signal using a bandpass filter to generate a filtered displacement signal;

generating a heartbeat rate estimate based on the filtered displacement signal;

determining a heartbeat rate prediction and an uncertainty prediction based on the track;

determining whether the heartbeat rate estimate is an outlier based on the uncertainty prediction; and

when the heartbeat rate estimate is determined to be not an outlier,

updating the track based on the heartbeat rate estimate, and

adjusting a lower cutoff frequency of the bandpass filter or a higher cutoff frequency the bandpass filter based on the uncertainty prediction.

19. A method for tracking a vital rate indicative of a heartbeat rate of a human target using a track of a Kalman filter, the method comprising:

receiving reflected radar signals;

generating a displacement signal indicative of a displacement of a target based on the reflected radar signals;

filtering the displacement signal using a bandpass filter to generate a filtered displacement signal;

generating a heartbeat rate estimate based on the filtered displacement signal;

determining a heartbeat rate prediction and an uncertainty prediction based on the track:

determining a gating region using an ellipsoidal gating function based on the uncertainty prediction;

determining that the heartbeat rate estimate is an outlier when a result of the ellipsoidal gating function is higher than a predetermined threshold;

determining that the heartbeat rate estimate is not an outlier when the result of the ellipsoidal gating function is lower than the predetermined threshold; and

when the heartbeat rate estimate is determined to be not an outlier,

updating the track based on the heartbeat rate estimate, and

updating a setting of the bandpass filter based on the updated track.

20. A method for tracking a vital rate indicative of a heartbeat rate of a human target using a track of a Kalman filter, the method comprising:

receiving reflected radar signals;

generating a displacement signal indicative of a displacement of a target based on the reflected radar signals;

filtering the displacement signal using a bandpass filter to generate a filtered displacement signal;

generating a heartbeat rate estimate based on the filtered displacement signal;

determining a heartbeat rate prediction and an uncertainty prediction based on the track;

determining whether the heartbeat rate estimate is an outlier based on the uncertainty prediction;

when the heartbeat rate estimate is determined to be not an outlier,

updating the track based on the heartbeat rate estimate, and

updating a setting of the bandpass filter based on the updated track; and

when the heartbeat rate estimate is determined to be an outlier:

incrementing a counter, and

increasing a pass-band of the bandpass filter when the counter reaches a count of M, wherein M is a positive integer greater than 1.