Cadence Detection FIELD OF INVENTION
[0001] The present invention relates to portable media devices and, more particularly, to method for detecting cadence of body motion.
BACKGROUND OF THE INVENTION
[0002] Characterizing body movements can be useful when applied to portable electronic devices such as cellular phones, game controllers, and portable computers.
These devices already incorporate sensors such as accelerometers and gyroscopes and have ample processing power.
The applications range from synchronizing media output for pleasure or for exercising, and to security or safety alerts.
[0003] Typical human activity involves walking, jogging and running. One attempt to obtain the cadence of this motion includes using a sensor embedded in a shoe detecting the repetitive motion of the leg. While it works it requires batteries and a special shoe and would therefore be an advantage to derive useful information from the sensors which are available in the portable media device. Furthermore, an alternative method may provide useful information for body motions other than running.
[0004] An example of the method to obtain cadence information from the accelerometer embedded in a portable electronic device is described in US Patent 7,457,719.
The method is based on identification of gravitational influence on rolling averages of acceleration over a sample periond and on generating a dominant axis based upon the gravitational influence.
However, the problem is that the accuracy of the method is compromised by the failure to detect rotation of the device and by lack of calibration for the global acceleration.
SUMMARY OF THE INVENTION
[0004] It is an object of the present invention to provide a method for detecting cadence of body motion from the signal of a sensor incorporated in the portable electronic device independent of the orientation, rotation or global acceleration of the device.
[0005] The invention relates, in one embodiment, to a method for detecting a cadence value of a body motion by measuring a signal from a sensor registering body motion of the user over a sample period. Cadence value is defined as a frequency of repeated set of body movements. The signal is transformed to obtain amplitude versus frequency information over a desired first range of frequencies. A low end cut-off value is then applied to the amplitudes in this range of frequencies. If a maximum amplitude frequency is detected in the first range of frequencies a second range of frequencies is created starting at the frequency set by s multiplier of the maximum amplitude frequency of the first range. A low end cut-off value is applied to the amplitudes of the second range of frequencies, the value being different than the cut-off value used in the first range. The maximum amplitude frequency from the first range is used as cadence value only if not detected in the second range. If maximum amplitude frequency is detected in both ranges the maximum from the second range is used as a cadence value.
[0006] The principle of detecting a cadence value described above can be extended by creating three or more ranges within the measured first range of frequencies. Every additional frequency range is created using the multiplier of the maximum of the previous range and applying a new cut-off value. A maximum amplitude frequency detected in the last range is applied as a cadence value.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] These and other features of the invention will become more apparent from the following description in which reference is made to the appended drawings wherein:
[0008] Figure 1 illustrates an exemplary motion graph displaying amplitude versus frequency transformation of a signal measured by an accelerometer;
[0009] Figure 2 illustrates exemplary motion graphs for a case when the same type of body motion intensifies from walking, jogging to running;
[0010] Figure 3 illustrates application of the method of detecting cadence of the body motion using two frequency ranges;
[0011] Figure 4 illustrates application of the method of detecting cadence of the body motion using three frequency ranges;
DETAILED DESCRIPTION
[0012] The following description is of a preferred embodiment.
[0013] A typical portable media device is measuring acceleration of the device in three coordinate directions relative to the orientation of the device. Since the device is located on the body of the user either firmly attached or loosely placed it changes its orientation while the user is engaged in a physical activity. If a vector of acceleration is calculated using information measured from the three coordinate directions it would denote a dominant acceleration of the body independent of the orientation of the device. When acceleration vector measurements are sampled over time sufficient information is obtained to perform amplitude versus frequency transformation allowing assessment of amplitudes at frequencies which are related to movements of individual body parts. Figure 1 shows one example of such transformation.
The peaks 2,3,4 and 5 of the curve 1 are caused by parts of the body moving at a particular frequency. Some of the peak frequencies can be more indicative of the general rhythm of the body than others. The selection could be subjective preference or the criterion can be that the frequency of the body part which is closer to the frequency intended application is desired. For example, in running frequency of leg movement is generally closer to tempo of music than frequency of the arm movement. However, from the graph shown in Figure 1 it is not obvious which peak belongs to the leg frequency.
[0014] When the motion of the body intensify while performing the same type of activity the amplitudes and the frequencies generated by the body parts change. For example, when walking progresses into jogging and running, the frequency of movements typically rises but the associated amplitudes increase or decrease. Figure 2 shows for walking the maximum peak B
associated with leg movement since it is obviously caused by the highest acceleration in this mode of body movement. While running, however, the amplitude peak A of the arm frequency may be higher than leg frequency peak B.
[0015] The movement frequency of the desired body part is used for synchronization of media as the cadence value. In order to obtain the cadence value reliably at any intensity of a particular body activity the set of rules is applied to the motion graph as shown in Figure 3. First, a cut-off amplitude value 6 is applied over the whole range 7 of the frequencies. The purpose is to discard amplitudes which are arbitrary. For example, just standing still or moving slowly the accelerometer may register peaks at frequencies unrelated clearly repetitive body movements. If a distinctive maximum amplitude 8 is found in the range 7 then a new evaluating range of frequencies 9 is created. This range starts at a distance 10 from the peak 8.
The distance 8 is determined using a multiplier which is determined by trial and error. Then a cut-off amplitude value 11 is applied over the range 9. The purpose of the cut-off value 11 is the same as the purpose for value 6. If a maximum amplitude frequency peak 12 is found in the range 9 it is used as a cadence value. If no amplitude frequency peak is found in the range 9 then peak frequency 8 is used as the cadence value.
[0016] The principle explained above can be expanded by applying more evaluating ranges to the same motion graph. Figure 4 shows using additional range 13 starting at distance 14 from the peak 12. A lower cut-off value 15 is shown in this example as compared to the value 11 of the previous range 9. If a maximum amplitude peak value 16 is found in the range 13 it is used as a cadence value.
[0017] In the foregoing specification, the invention has been described with reference to specific exemplary embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention as set forth in the appended claims.