Energy efficient mobile context collection

• US 20160282156A1
• Filed: 03/23/2015
• Published: 09/29/2016
• Est. Priority Date: 03/23/2015
• Status: Active Application

• Current Assignee: Incoming Pty Ltd
• Original Assignee: Incoming Pty Ltd
• Inventors: Ott, Maximilian, Petander, Lars Henrik
• Granted Patent: US 10,267,661 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01D 18/00 Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
• CPC Class Codes: H04W 4/02 Services making use of location information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services

Abstract

This disclosure relates to determining a context of a mobile device. A processor receives from one or more first sensors first sensor data in relation to the context. The processor then selects from multiple second sensors one or more selected sensors based on the first sensor data and based on an energy cost value associated with each of the multiple second sensors, the energy cost value being indicative of energy consumed when requesting data from each of the multiple second sensors. Next, the processor receives from the one or more selected sensors second sensor data and determines the context of the mobile device based on the first sensor data and the second sensor data. This reduces the overall energy consumption and therefore extends the battery life of the mobile device while still providing an accurate context determination.

Claims

1. A method for determining a context of a mobile device, the method comprising:

receiving from one or more first sensors first sensor data in relation to the context;

selecting from multiple second sensors one or more selected sensors based on the first sensor data and based on an energy cost value associated with each of the multiple second sensors, the energy cost value being indicative of energy consumed when requesting data from each of the multiple second sensors;

receiving from the one or more selected sensors second sensor data; and

determining the context of the mobile device based on the first sensor data and the second sensor data.

2. The method of claim 1, wherein the first sensor data comprises an identification of the one or more first sensors.

3. The method of claim 1, further comprising determining an information gain associated with each of the second set of sensors based on the first sensor data, wherein selecting from multiple second sensors one or more selected sensors comprises selecting from multiple second sensors one or more selected sensors based on the information gain.

4. The method of claim 3, wherein determining the information gain comprises:

evaluating a model using the first sensor data as input data, the model comprising parameters related to the multiple first sensors and the multiple second sensors; and

determining the information gain based on an output of the model.

5. The method of claim 3, wherein selecting from multiple second sensors one or more selected sensors comprises selecting from multiple second sensors one or more selected sensors based on an information threshold.

6. The method of claim 5, further comprising determining a weighted sum based on the first sensor data, wherein selecting from multiple second sensors one or more selected sensors comprises selecting one or more second sensors that are associated with a minimal energy cost while at the same time being associated with an information gain greater than the distance of the weighted sum from the information threshold.

7. The method of claim 5 or 6, wherein the information threshold is based on a number of base learners that have reached a decision in an ensemble model and how many base learners still need to reach a decision before the ensemble model can reach a majority vote such that the one or more selected sensors are the sensors which allow the ensemble to reach the decision at an expected lowest energy cost.

8. The method of claim 7, wherein the base learners comprise a random forest or boosted trees.

9. The method of claim 1, wherein receiving the second sensor data comprises:

determining a point in time to retrieve the second sensor data at an optimised energy cost; and

receiving the second sensor data at the point in time.

10. The method of claim 1, wherein receiving from the one or more selected sensors second sensor data comprises receiving from more than one sensors second sensor data in parallel.

11. The method of claim 10, wherein determining the context comprises evaluating a context model using the second sensor data from each of the more than one second sensors sequentially as input data to the context model.

12. The method of claim 1, wherein selecting from multiple second sensors one or more selected sensors comprises selecting from multiple second sensors one or more selected sensors based on a sensor access model.

13. The method of claim 12, wherein the sensor access model is a decision tree.

14. The method of claim 12, wherein the sensor access model is a cascaded learning method.

15. The method of claim 1, wherein the one or more first sensors are activated independently from selecting the one or more selected sensors and determining the context of the mobile device.

16. The method of claim 1, further comprising excluding one or more excluded sensors from the one or more second sensors based on the first sensor data.

17. The method of claim 16, wherein excluding the one or more excluded sensors comprises determining the excluded sensors based on one or more of:

domain knowledge,

database inference, and

context data associated with the mobile device.

18. The method of claim 1, wherein the one or more first sensors are associated with a communications network to which the mobile device is communicatively coupled.

19. The method of claim 1, wherein

selecting the one or more selected sensors comprises sending sensor selection data indicative of the one or more selected sensors to a mobile device,

receiving the second sensor data comprises receiving the second sensor data from the mobile device over a data network, and

determining the context comprises sending context data indicative of the context to the mobile device.

20. The method of claim 1, wherein the method is performed by a first process on the mobile device and the one or more first sensors are activated by a second process on the mobile device, the second process being different to the first process.

21. The method of claim 1, further comprising waiting for the first sensor data.

22. The method of claim 1, wherein selecting from multiple second sensors one or more selected sensors comprises selecting from multiple second sensors one or more selected sensors based on an age of the first sensor data.

23. The computer-implemented method of claim 1, wherein the first sensor data is indicative of one or more of:

a charging event of a battery of the mobile device; and

a current cyclical time.

24. The computer-implemented method of claim 1, wherein the first sensor data is indicative of a connectivity status change in relation to one or more of:

a Bluetooth device;

a Wifi hotspot; and

a cellular network cell.

25. The method of claim 1, further comprising one or more of:

downloading content item based on the context; and

displaying a notification based on the context.

26. The method of claim 1, further comprising training a context model based on the first sensor data and the second sensor data, wherein determining the context comprises evaluating the context model.

27. A non-transitory computer readable medium with an executable program stored thereon that when executed causes a computer to perform the method of claim 1.

28. A computer system for determining a context of a mobile device, the computer system comprising:

an input data port;

a processor to

receive through the input data port from one or more first sensors first sensor data in relation to the context;

select from multiple second sensors one or more selected sensors based on the first sensor data and based on an energy cost value associated with each of the multiple second sensors, the energy cost value being indicative of energy consumed when requesting data from each of the multiple second sensors;

receive through the input data port from the one or more selected sensors second sensor data; and

determine the context of the mobile device based on the first sensor data and the second sensor data.