Method for the estimation of the heart-rate and corresponding system

1. A method, comprising:

selecting heart-beat data blocks based on a received heart-beat signal;

selecting acceleration data blocks based on one or more received acceleration signals;

converting the selected heart-beat data blocks to heart-beat frequency domain data blocks;

converting the selected acceleration data blocks to acceleration frequency domain data blocks;

performing motion compensation in the frequency domain based on the converted frequency domain data blocks;

generating an estimated heart rate based on the motion compensation;

generating a predicted estimate based on the acceleration frequency domain data blocks; and

generating a heart-rate signal based on the estimated heart rate and the predicted estimate, wherein a linear function is used to generate the predicted estimate, and the linear function is updated based on the generated heart-rate signal.

2. The method of claim 1 , comprising generating the received heart-beat signal using photoplethysmography on a wrist.

3. The method of claim 1 wherein said motion compensation includes:

multiplying the acceleration frequency domain data blocks by respective scalar weights; and

subtracting the weighted acceleration frequency domain data blocks from the heart-beat frequency domain data blocks.

4. The method of claim 1 , comprising:

generating a compensated heart-beat signal; and

generating the estimated heart rate based on the compensated heart-beat signal.

5. The method of claim 4 , comprising:

computing a non-linear predictor value based on the acceleration frequency domain data blocks; and

generating the predicted estimate as a linear function of the predictor value.

6. The method of claim 5 wherein the generating the heart-rate signal comprises selecting one of the estimated heart rate and the predicted estimate.

7. The method of claim 5 , comprising:

averaging a plurality of estimated heart rates estimated at different times.

8. The method of claim 6 , comprising:

updating said linear function by storing a current estimated heart rate and predictor value when the estimated heart rate is selected, obtaining a sequence of observation points; and

calculating the linear function as a regression over said sequence of observation points.

9. The method of claim 5 , comprising performing an aggregation operation over said sequence of observation points.

10. The method of claim 1 , comprising:

filtering the received heart-beat signal and selecting the heart-beat data blocks based on the filtered heart-beat signal;

filtering the received one or more acceleration signals and selecting the acceleration data blocks based on the filtered one or more acceleration signals, the filtering including low-pass filtering and high-pass filtering, the high-pass filtering using a chain of infinite impulse response filters.

11. The method of claim 1 , comprising:

generating a data block based on the received heart-beat signal;

determining whether the generated data block satisfies a testing condition;

when the generated data block satisfies the testing condition, converting the generated data block and data blocks based on the received one or more acceleration signals which are time-aligned with said generated data block to frequency domain data blocks; and

when the generated data block does not satisfy the testing condition, discarding the generated data block and generating a new data block that is time-shifted relative to the previously generated data block.

12. The method of claim 4 wherein the estimating the heart rate comprises identifying a frequency of the compensated heart-beat signal.

13. The method of claim 1 , comprising:

interleaving a plurality of Fast Fourier Transform operations.

14. The method of claim 1 , comprising:

transmitting the generated heart-rate signal.

15. A device, comprising:

an interface configured to receive a heart-beat signal and one or more acceleration signals: and

signal processing circuitry coupled to the interface and configured to:

select heart-beat data blocks based on the heart-beat signal;

select acceleration data blocks based on the one or more acceleration signals;

convert the selected heart-beat data blocks to heart-beat frequency domain data blocks;

convert the selected acceleration data blocks to acceleration frequency domain data blocks;

perform motion compensation in the frequency domain based on the converted frequency domain data blocks;

generate an estimated heart rate based on the motion compensation;

generate a predicted estimate based on the acceleration frequency domain data blocks; and

generate a heart-rate signal based on the estimated heart rate and the predicted estimate, wherein a linear function is used to generate the predicted estimate, and the linear function is updated based on the generated heart-rate signal.

16. The device of claim 15 , comprising:

one or more light sources; and

one or more optical sensors configured to generate the heart-beat signal.

17. The device of claim 15 wherein the signal processing circuitry is configured to:

multiply the acceleration frequency domain data blocks by respective scalar weights; and

subtract the weighted acceleration frequency domain data blocks from the heart-beat frequency domain data blocks.

18. The device of claim 15 wherein the signal processing circuitry is configured to:

generate a compensated heart-beat signal; and

generate the estimated heart rate based on the compensated heart-beat signal.

19. The device of claim 18 wherein the signal processing circuitry is configured to:

compute a non-linear predictor value based on the acceleration frequency domain data blocks; and

generate the predicted estimate as a linear function of the predictor value.

20. The device of claim 19 wherein the signal processing circuitry is configured to select one of the estimated heart rate and the predicted estimate.

21. The device of claim 19 wherein the signal processing circuitry is configured to: average a plurality of estimated heart rates estimated at different times.

22. The device of claim 20 wherein the signal processing circuitry is configured to:

update said linear function by storing a current estimated heart rate and predictor value when the estimated heart rate is selected, obtaining a sequence of observation points; and

calculate the linear function as a regression over said sequence of observation points.

23. The device of claim 15 wherein the signal processing circuitry comprises:

a plurality of low-pass filters and a plurality of high-pass filters configured to filter the received signals.

24. The device of claim 15 wherein the signal processing circuitry is configured to:

generate a data block based on the received heart-beat signal;

determine whether the generated data block satisfies a testing condition;

when the generated data block satisfies the testing condition, convert the generated data block and data blocks based on the received one or more acceleration signals which are time-aligned with said generated data block to frequency domain data blocks; and

when the generated data block does not satisfy the testing condition, discard the generated data block and generate a new data block that is time-shifted relative to the previously generated data block.

25. The device of claim 15 , comprising:

a transmitter coupled to the signal processing circuitry and configured to transmit the generated heart-rate signal.

26. A system, comprising:

an optical sensor, which, in operation, generates a heart-beat signal;

an accelerometer, which, in operation, generates one or more acceleration signals; and

signal processing circuitry communicatively coupled to the optical sensor and the accelerometer, wherein the signal processing circuitry, in operation:

selects heart-beat data blocks based on the heart-beat signal;

selects acceleration data blocks based on the one or more acceleration signals;

converts the selected heart-beat data blocks to heart-beat frequency domain data blocks;

converts the selected acceleration data blocks to acceleration frequency domain data blocks;

performs motion compensation in the frequency domain based on the converted frequency domain data blocks;

generates an estimated heart rate based on the motion compensation;

generates a predicted estimate based on the acceleration frequency domain data blocks; and

generates a heart-rate signal based on the estimated heart rate and the predicted estimate, wherein a linear function is used to generate the predicted estimate, and the linear function is updated based on the generated heart-rate signal.

27. The system of claim 26 , comprising:

an integrated circuit including the signal processing circuitry.

28. The system of claim 27 wherein the integrated circuit includes the optical sensor and the accelerometer.

29. The system of claim 26 , comprising:

a transmitter configured to transmit the generated heart-rate signal.

30. The system of claim 26 , comprising a display.

31. A non-transitory computer-readable medium having contents which configure a heart-rate monitoring device to perform a method, the method comprising:

selecting heart-beat data blocks based on a received heart-beat signal;

selecting acceleration data blocks based on one or more received acceleration signals;

converting the selected heart-beat data blocks to heart-beat frequency domain data blocks;

converting the selected acceleration data blocks to acceleration frequency domain data blocks;

performing motion compensation in the frequency domain based on the converted frequency domain data blocks;

generating an estimated heart rate based on the motion compensation;

generating a predicted estimate based on the acceleration frequency domain data blocks; and

generating a heart-rate signal based on the estimated heart rate and the predicted estimate, wherein a linear function is used to generate the predicted estimate, and the linear function is updated based on the generated heart-rate signal.

32. The medium of claim 31 wherein the method comprises:

multiplying the acceleration frequency domain data blocks by respective scalar weights; and

subtracting the weighted acceleration frequency domain data blocks from the heart-beat frequency domain data blocks.

33. The medium of claim 32 wherein the method comprises:

generating a compensated heart-beat signal; and

generating the estimated heart rate based on the compensated heart-beat signal.

34. The medium of claim 33 wherein the method comprises:

computing a non-linear predictor value based on the acceleration frequency domain data blocks; and

generating the predicted estimate as a linear function of the predictor value.

35. The medium of claim 31 wherein the method comprises:

transmitting the generated heart-rate signal.