Time and frequency domain based activity tracking system

• US 10,653,339 B2
• Filed: 04/08/2015
• Issued: 05/19/2020
• Est. Priority Date: 04/29/2014
• Status: Active Grant

• Current Assignee: NXP B.V. (NXP Semiconductors NV)
• Original Assignee: NXP B.V. (NXP Semiconductors NV)
• Inventors: Gaddam, Vasanth, Zhang, Yifeng, Zhang, Jie, Wu, Yuanwei, Wang, Guanqing
• Primary Examiner(s): Sengdara, Vongsavanh
• Application Number:
  US14/681,438
• Publication Number:
  US 20160296144A1
• Time in Patent Office: 1,868 Days
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/1123 Discriminating type of movement, e.g. walking or running A61B5/1116, A61B5/112 take precedence
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G01C 21/20 Instruments for performing navigational calculations G01C21/24, G01C21/26 take precedence
• CPC Class Codes: G01C 22/006 Pedometers
• CPC Class Codes: G01P 15/00 Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: H01L 21/32139 using masks
• CPC Class Codes: H01L 29/6653 using the removal of at least part of spacer, e.g. disposable spacer
• CPC Class Codes: H01L 29/66545 using a dummy, i.e. replacement gate in a process wherein at least a part of the final gate is self aligned to the dummy gate
• CPC Class Codes: H01L 29/6656 using multiple spacer layers, e.g. multiple sidewall spacers
• CPC Class Codes: H01L 29/66795 with a horizontal current flow in a vertical sidewall of a semiconductor body, e.g. FinFET, MuGFET
• CPC Class Codes: H01L 29/7851 with the body tied to the substrate

Abstract

A method and apparatus for activity tracking based on time domain and frequency domain processing are disclosed. Embodiments according to the present invention are used to improve the accuracy of activity detection and step counting. The activity tracking starts from sample collection to generate 3-D accelerometer data. By pre-processing, the 3-D accelerometer data is calibrated and filtered. Then, the dominant component is calculated and statistical attributes or features used for activity detection are extracted. The statistical attributes are derived from time domain sensor data, frequency domain transformed data, or both. A classifier is developed using representative training data set. The activity detector determines the current activity status based on the statistical attributes and the classifier. To further refine the activity, post-processing is performed on the activity status.

Claims

1. A system for activity tracking, the system comprising:

a 3-dimensional (3-D) accelerometer configured to sense activity in each of three orthogonal axes being X-axis, Y-axis and Z-axis, and generating accelerometer data associated with the three axes;

a pre-processor coupled to the 3-D accelerometer and including a filter circuit configured and arranged to remove noise from the accelerometer data and, therefrom produce filtered data including vertical and horizontal component data and accelerometer data associated with each of the three axes; and

a computer processor configured and arranged with previously-obtained statistical data indicative of a plurality of attributes of each different activity classes characterizing movement of a user and with programming instructions which when executed, the computer processor performs operations to identify, independent of orientation of the 3-D accelerometer, one of the different activity classes corresponding to produced accelerometer data, the operations including the following;

deriving, from the filtered data, values associated with activity-indicative attributes including mean value, time domain energy of filtered data, and variance of the vertical and horizontal component data,

processing and comparing each of the values of the activity-indicative attributes with the previously-obtained statistical data to classify the produced accelerometer data as corresponding to a selected current activity from among the different activity classes, and

tracking and characterizing the tracked data as the selected current activity of one of the different activity classes, wherein the selected current activity corresponds with the movement of the user.

2. The system of claim 1, wherein the filter is further configured and arranged to remove direct current (DC) component from the accelerometer data.

3. The system of claim 1, wherein the computer processor is further configured to, in response to tracking and characterizing the tracked data as the selected current activity, log activity of the user associated with the different activity classes over a period of time and provide performance data indicative of the logged activity including the tracked data characterized as the selected current activity.

4. The system of claim 1, wherein the computer processor is further configured to determine which of the vertical component and the horizontal component is dominant based on an amount of energy indicated the vertical component and the horizontal component and to characterizing the tracked data as the selected current activity based on a variance of the dominant one of the vertical component and the horizontal component.

5. The system of claim 1, wherein the pre-processor is configured to produce the filtered data including vertical and horizontal component data and accelerometer data associated with each of the three axes independent of the orientation of the 3-D accelerometer.

6. A non-transitory computer readable media including programming instruction which when executed by a computer processor performs the following operations:

continuously sensing, using a 3-dimensional (3-D) accelerometer, activity in each of three orthogonal axes being X-axis, Y-axis and Z-axis corresponding with movement of a user;

pre-processing accelerometer data which is output from the 3-D accelerometer, wherein the pre-processing includes causing the data which is output from the 3-D accelerometer to be filtered by removing noise and removing a direct current (DC) component from an output of the 3-D accelerometer;

deriving, after the pre-processing, a vertical component and a horizontal component of the output of the 3-D accelerometer, wherein the vertical component is proportional to a vector inner product between a mean vector of a plurality of vector samples along the three axes and each vector corresponding to a first difference between each of the plurality of vector samples and the mean vector, wherein the horizontal component is related to a norm of a second difference between the first difference and the mean vector scaled by the vertical component;

determining which of the vertical component and the horizontal component indicates an amount of energy that is greater than an amount of energy indicated by the other of the vertical component and the horizontal component; and

determining, tracking and characterizing an activity type from a plurality of activity types characterizing the movement of the user, independent of orientation of the 3-D accelerometer and based on statistical data corresponding to the output of the 3-D accelerometer and the vertical component or the horizontal component having the greater amount of energy.

7. The computer readable media of claim 6, further including determining the activity type based on a step counting, wherein the step counting is calculated based at least partly on the vertical component or the horizontal component having the greater amount of energy.

8. The computer readable media of claim 6, wherein the statistical data includes:

at least one of a time domain mean value, a variance, a signal energy of the vertical component or the horizontal component having the greater amount of energy;

a first set of statistical data generated based on the pre-processing; and

a second set of statistical data generated based on the determining whether the vertical component or the horizontal component has the greater amount of energy.

9. The computer readable media of claim 6, wherein determining the activity type is further based on cumulative distribution for first values of the statistical data, wherein the cumulative distribution is divided into multiple sections associated with first values of the statistical data and second values of a plurality of statistical features of a particular activity, and wherein each section corresponds to a ratio range of the first values to the second values.

10. The computer readable media of claim 6, wherein determining the activity type is further based on cumulative distribution for first values of the statistical data, wherein the cumulative distribution is divided into multiple sections associated with first values of the statistical data and second values of a plurality of statistical features of a particular activity, and wherein each section corresponds to a ratio range of the first values to the second values as follows:

great than 0 and equal to or less than 5%, great than 5% and equal to or less than 10%, great than 10% and equal to or less than 90%, great than 90% and equal to or less than 95%, and great than 95% and equal to or less than 99%.

11. The computer readable media of claim 6, wherein determining, tracking and characterizing the activity type from the plurality of activity types further includes determining which of the vertical component or the horizontal component are a dominant component responsive one of the vertical component or the horizontal component having the greater amount of energy, and characterizing the activity type based on a variance of the dominant component.

12. The computer readable media of claim 6, wherein in response to determining, tracking and characterizing the activity type, the operations further include logging activity of the user associated with the activity type and providing performance data indicative of the logged activity to the user.

13. A computer readable media including programming instructions which when executed by a processor performs an operation for determining an activity from a set of multiple possible activities, the operation includes:

sensing, using a 3-dimensional (3-D) accelerometer, activity in each of three orthogonal axes being X-axis, Y-axis and Z-axis corresponding with movement of a user;

receiving raw data from the 3-D accelerometer;

causing the raw data to be filtered by removing noise and removing a direct current (DC) component from the raw data;

calculating a time domain mean value and a time domain energy of the filtered raw data;

estimating a component in the filtered raw data having a greatest amount of energy; and

determining, tracking and characterizing the activity characterizing the movement of the user, independent of orientation of the 3-D accelerometer and based on the estimated component and the time domain mean value and the time domain energy.

14. The computer readable media of claim 13, wherein the operation further includes calculating the time domain mean value prior to the estimating.

15. The computer readable media of claim 13, wherein determining, tracking and characterizing the activity characterizing the movement of the user, further includes identifying a variance of the estimated component, and determining the activity based on the variance of the estimated component and the time domain mean value and the time domain energy.

16. The computer readable media of claim 13, the operation further including determining, tracking and characterizing activity characterizing movement of the user corresponding to a plurality of activities from the set of multiple possible activities over time, logging activity of the user associated with the plurality of activities and providing performance data indicative of the logged activity to the user.