Method and device for retrieving biomechanical parameters of a stride

1. A method for determining biomechanical parameters of the stride of a runner, using an accelerometer device on one wrist, on one upper arm, on the head or on one shoe, comprising the steps of

measuring an initial sequence of acceleration data in at least the vertical direction using said accelerometer device, said initial sequence of acceleration data depending on the motion of the runner's center of mass, and on extra-motions of the accelerometer device relative to the runner's center of mass;

identifying in said initial sequence of acceleration data at least one frequency component caused by relative motion of the accelerometer device relative to the runner center of mass;

attenuating said frequency component, so as to attenuate the component of acceleration due to said extra motions, so as to determine a modified sequence of acceleration data corresponding to the one that would be measured at the runner's center of mass;

determining said biomechanical parameters of the stride from said modified sequence of acceleration data.

2. The method of claim 1 , wherein said step of identifying comprises:

extracting the cadence directly from said initial sequence of acceleration data;

using said cadence for determining said frequency component.

3. The method of claim 2 , further comprising:

verifying that the cadence determined directly from said initial sequence of acceleration data corresponds to the frequency peak.

4. The method of claim 2 , wherein said step of identifying further comprises retrieving at least one harmonic peak from said cadence and/or from said most significant peak,

and wherein said step of attenuating comprises attenuating said harmonic peak.

5. The method of claim 4 , wherein frequencies between two attenuated harmonics are less attenuated than said harmonics.

6. The method of claim 2 , wherein said step of attenuating comprises attenuating frequencies at half the frequency of the cadence peak.

7. The method of claim 2 , wherein said step of attenuating comprises attenuating frequencies at some harmonics of the cadence peak.

8. The method of claim 2 , wherein said step of attenuating comprises attenuating the cadence peak by a first attenuation factor, and applying a greater attenuation to harmonics corresponding to the arm swing peak.

9. The method of claim 1 , wherein said step of identifying further comprises:

extracting a sample window from said initial sequence of acceleration;

performing a FFT of said sample window;

determining the most significant frequency peak from the transformed signal.

10. The method of claim 1 , further comprising:

extracting a sample window from said initial sequence of acceleration;

performing a FFT of said sample window, so as to compute a FFT signal;

inputting said FFT signal to a machine learning system, such as for example a neural network, in order to determine the cadence of the runner.

11. The method of claim 1 , performed in a wristwatch.

12. The method of claim 1 , wherein said step of measuring is performed in a wristwatch, while said steps of identifying, attenuating and determining are performed in a smartphone or in a computer.

13. The method of claim 1 , wherein said biomechanical components include at least one among:

flight time, contact time, lowering, elevation, undulation, asymmetry, regularity, reactivity, stiffness, take of angle, and/or landing angle.

14. The method of claim 1 , further comprising a step of using said at least one frequency component caused by relative motion of the accelerometer device relative to the runner center of mass, in order to determine the arm movement.

15. The method of claim 1 , said initial acceleration data being measured in a plurality of orthogonal directions, said attenuation being performed independently for each direction.

16. The method of claim 15 , wherein the attenuated frequencies are the same for each direction.

17. An accelerometer device for determining biomechanical parameters of the stride of a runner with a high accuracy, comprising:

means for fastening the device to one wrist, to one upper arm, to the head or to one shoe;

an accelerometer for measuring an initial sequence of acceleration data in at least the vertical direction, said initial sequence of acceleration data depending on the motion of the runner's center of mass, and on extra-motions of the accelerometer device relative to the runner's center of mass;

a processing system containing a processor and a memory containing a program code arranged for:

identifying in said initial sequence of acceleration data at least one frequency component caused by relative motion of the accelerometer device relative to the runner center of mass;

attenuating said frequency component, so as to attenuate the component of acceleration due to said extra motions, so as to determine a modified sequence of acceleration data corresponding to the one that would be measured at the runner's center of mass;

determining said biomechanical parameters of the stride from said modified sequence of acceleration data.

18. The accelerometer device of claim 17 , further comprising:

a battery;

a display.

19. A tangible computer product containing program code for causing a processor to execute the method of claim 1 when said code is executed.