Battery drain analysis and prediction for wireless devices

1. A method implemented by data processing apparatus, the method comprising:

extracting first attributes associated with a plurality of first applications;

collecting data samples from a plurality of wireless devices that are configured to execute one or more of the first applications, each of the data samples including (i) one or more application identifiers identifying one or more processes executing on a wireless device of the plurality of wireless devices during a time duration, and (ii) battery drain information associated with the time duration;

determining battery usage characteristics of the extracted first attributes based on correlations among at least a portion of the data samples;

extracting one or more second attributes associated with a second application;

identifying one or more of the extracted first attributes that correspond respectively to the one or more extracted second attributes as one or more identified first attributes;

using one or more of the determined battery usage characteristics that correspond respectively to the one or more identified first attributes to determine a battery usage profile for the second application;

receiving an identifier associated with the second application from a requesting wireless device; and

providing, based on the received identifier, the battery usage profile to the requesting wireless device.

2. The method of claim 1 , wherein determining the battery usage characteristics comprises generating statistical coefficients for the extracted first attributes, and wherein the battery usage profile for the second application is based on one or more of the statistical coefficients.

3. The method of claim 1 , wherein the battery usage profile includes a predicted battery drain rate.

4. A system comprising:

a storage device configured to store data samples from a plurality of wireless devices that are configured to execute one or more of a plurality of first applications, the data samples being associated with at least a portion of the one or more of the first applications, each of the data samples including (i) one or more application identifiers identifying one or more processes executing on a wireless device of the plurality of wireless devices during a time duration, and (ii) battery drain information associated with the time duration; and

a processor communicatively coupled with the storage device, configured to (i) extract first attributes associated with the plurality of first applications, (ii) determine battery usage characteristics of the first attributes based on correlations among at least a portion of the data samples, (iii) extract one or more second attributes associated with a second application, (iv) identify one or more of the first attributes that correspond respectively to the one or more second attributes as one or more identified first attributes, and (v) use one or more of the battery usage characteristics that correspond respectively to the one or more identified first attributes to determine a battery usage profile for the second application,

wherein the processor is configured to receive an identifier associated with the second application from a requesting wireless device, and provide, based on the identifier, the battery usage profile to the requesting wireless device.

5. The system of claim 4 , wherein the processor is configured to generate statistical coefficients for the first attributes, and wherein the battery usage profile for the second application is based on one or more of the statistical coefficients.

6. The system of claim 4 , wherein the battery usage profile includes a predicted battery drain rate.

7. A method implemented by data processing apparatus, the method comprising:

collecting data samples from a plurality of wireless devices that are configured to execute first applications that are collectively associated with first attributes, each of the data samples including (i) one or more application identifiers identifying one or more processes executing on a wireless device of the plurality of wireless devices during a time duration, and (ii) battery drain information associated with the time duration;

determining battery usage characteristics of the first attributes based on correlations among at least a portion of the data samples; and

extrapolating a battery usage profile for a second application based on one or more similarities between the second application and the first applications, the second application being associated with one or more second attributes, wherein the battery usage profile is based on one or more of the determined battery usage characteristics of the first attributes that correspond respectively to one or more of the one or more second attributes,

wherein determining the battery usage characteristics comprises generating statistical coefficients for the first attributes, wherein the battery usage profile for the second application is based on one or more of the statistical coefficients.

8. The method of claim 7 , comprising:

storing the determined battery usage characteristics in a database;

identifying one or more common attributes between the second application and the first applications; and

retrieving from the database, one or more battery usage characteristics for the one or more common attributes, wherein the battery usage profile is based on the one or more retrieved battery usage characteristics.

9. The method of claim 7 , wherein the battery usage profile includes a predicted battery drain rate.

10. The method of claim 7 , comprising:

receiving an identifier associated with the second application from a requesting wireless device; and

providing, based on the received identifier, the battery usage profile to the requesting wireless device.

11. The method of claim 7 , comprising:

determining a plurality of prediction outcomes for a plurality of battery usage prediction models, wherein the battery usage prediction models include a static model that is based on the battery usage profile;

generating a battery prediction result based on a weighting of the prediction outcomes; and

determining an estimated time to a depleted state of a battery of a wireless device based on the battery prediction result.

12. A system comprising:

a storage device configured to store data samples from a plurality of wireless devices that are configured to execute one or more of a plurality of first applications that are collectively associated with first attributes, each of the data samples including (i) one or more application identifiers identifying one or more processes executing on a wireless device of the plurality of wireless devices during a time duration, and (ii) battery drain information associated with the time duration; and

a processor communicatively coupled with the storage device, configured to (i) determine battery usage characteristics of the first attributes based on correlations among at least a portion of the data samples, and (ii) extrapolate a battery usage profile for a second application based on one or more similarities between the second application and the first applications, the second application being associated with one or more second attributes,

wherein the battery usage profile is based on one or more of the battery usage characteristics of the first attributes that correspond respectively to one or more of the one or more second attributes,

wherein the processor is configured to generate statistical coefficients for the first attributes, and wherein the battery usage profile for the second application is based on one or more of the statistical coefficients.

13. The system of claim 12 , wherein the processor is configured to

store the battery usage characteristics in a database,

identify one or more common attributes between the second application and the first applications, and

retrieve from the database, one or more battery usage characteristics for the one or more common attributes, wherein the battery usage profile is based on the one or more retrieved battery usage characteristics.

14. The system of claim 12 , wherein the battery usage profile includes a predicted battery drain rate.

15. The system of claim 12 , wherein the processor is configured to

receive an identifier associated with the second application from a requesting wireless device, and

provide, based on the identifier, the battery usage profile to the requesting wireless device.

16. The system of claim 12 , wherein the processor is configured to

determine a plurality of prediction outcomes for a plurality of battery usage prediction models, wherein the battery usage prediction models include a static model that is based on the battery usage profile,

generate a battery prediction result based on a weighting of the prediction outcomes, and

determine an estimated time to a depleted state of a battery of a wireless device based on the battery prediction result.

17. The system of claim 12 , comprising:

a wireless device comprising a battery, and configured to (i) determine a plurality of prediction outcomes for a plurality of battery usage prediction models, wherein the battery usage prediction models include a static model that is based on the battery usage profile, (ii) generate a battery prediction result based on a weighting of the prediction outcomes, and (iii) determine an estimated time to a depleted state of the battery based on the battery prediction result.

18. A method implemented by data processing apparatus comprising:

collecting data samples from a plurality of wireless devices that are configured to execute first applications that are collectively associated with first attributes, each of the data samples including (i) one or more application identifiers identifying one or more processes executing on a wireless device of the plurality of wireless devices during a time duration, and (ii) battery drain information associated with the time duration;

determining battery usage characteristics of the first attributes based on correlations among at least a portion of the data samples; and

determining a battery usage profile for a second application, the second application being different from the first applications, the second application being associated with second attributes, wherein determining the battery usage profile for the second application comprises: identifying common attributes that are in common between the second attributes of the second application and the first attributes of the first applications, and extrapolating the battery usage profile for the second application based on the battery usage characteristics that correspond to the common attributes.

19. The method of claim 18 , comprising:

extracting the first attributes from the first applications, wherein extracting the first attributes comprises assigning an attribute to at least one of the first applications based on a determination of whether a device resource is used by the at least one of the first applications.

20. The method of claim 18 , comprising:

extracting the first attributes from the first applications, wherein extracting the first attributes comprises assigning an attribute to at least one of the first applications based on a determination of whether a library is used by the at least one of the first applications.

21. A system comprising:

a storage device configured to store data samples from a plurality of wireless devices that are configured to execute one or more of a plurality of first applications, the first applications being collectively associated with first attributes, the data samples being associated with at least a portion of the one or more of the first applications, each of the data samples including (i) one or more application identifiers identifying one or more processes executing on a wireless device of the plurality of wireless devices during a time duration, and (ii) battery drain information associated with the time duration; and

a processor communicatively coupled with the storage device, configured to (i) determine battery usage characteristics of the first attributes based on correlations among at least a portion of the data samples, and (ii) determine a battery usage profile for a second application, the second application being different from the first applications, the second application being associated with second attributes,

wherein the processor is configured to identify common attributes that are in common between the second attributes of the second application and the first attributes of the first applications, and extrapolate the battery usage profile for the second application based on the battery usage characteristics that correspond to the common attributes.

22. The system of claim 21 , wherein the processor is configured to extract one or more of the first attributes by at least assigning an attribute to at least one of the first applications based on a determination of whether a device resource is used by the at least one of the first applications.

23. The system of claim 21 , wherein the processor is configured to extract one or more of the first attributes by at least assigning an attribute to at least one of the first applications based on a determination of whether a library is used by the at least one of the first applications.