Systems and methods of power management based on user behavior

• US 9,965,725 B1
• Filed: 12/01/2014
• Issued: 05/08/2018
• Est. Priority Date: 04/18/2010
• Status: Active Grant

• Current Assignee: Aptima Inc.
• Original Assignee: Aptima Inc.
• Inventors: Levchuk, Georgiy, Schurr, Nathan, Hering, Darby, Zakin, Mitch
• Primary Examiner(s): Lee, Thomas
• Assistant Examiner(s): Chang, Vincent
• Application Number:
  US14/557,435
• Time in Patent Office: 1,254 Days
• Field of Search: None
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode
• CPC Class Codes: G06F 1/3206 Monitoring of events, devices or parameters that trigger a change in power modality
• CPC Class Codes: G06F 1/3228 Monitoring task completion, e.g. by use of idle timers, stop commands or wait commands
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: H04B 17/391 Modelling the propagation channel
• CPC Class Codes: H04W 52/223 predicting future states of the transmission

Abstract

A method for power management comprising inferring a user behavior from an action, inferring a mission state from the action and an event, forecasting a forecasted action from the user behavior and the mission state and outputting an instruction to modify a power resource allocation based on the forecasted action. A processor based assembly for power management of at least one device comprising a means to infer a user behavior from an action, a means to infer a mission state from the action and an event, a means to forecast and a means to plan power management from the inferred information. In some embodiments, the systems and methods apply pattern recognition algorithms and pattern learning algorithms to manage the power allocation to power consuming devices.

Claims

1. A method for power management of a battery powered processor based device, said method comprising:

inferring a user behavior from an action;

the action comprising a first and a second action;

the action further comprising an application use on a processor based device;

inferring a mission state of a mission from the action;

the mission comprises a plurality of tasks and each task reflecting a degree of task completion;

forecasting, with a processor, a forecasted action from the user behavior and the mission state; and

outputting an instruction to modify a power resource allocation for the processor based device based on the forecasted action wherein the instruction to modify a power resource allocation comprises modifying a power mode of an application.

2. The method of claim 1 wherein:

the application use comprises the user selecting an application; and

the inferring a user behavior from an action comprises automatically determining a probability of user behavior given the application use.

3. The method of claim 1 wherein the inferring a mission state from the action comprises automatically determining a probability of a mission state from the action and with a pattern recognition algorithm.

4. The method of claim 1 wherein:

the application use representing the user selecting an application; and

the inferring a mission state from the action and comprises automatically determining a mission state based on the application use.

5. The method of claim 1 wherein:

the inferring a mission state from the action comprises automatically determining a joint probability of a mission state according to a probabilistic attributed graph matching algorithm comprising an app use generation factor, a task relation match factor and a mission state prior factor.

6. The method of claim 1 wherein:

the action comprises the user selecting an application; and

the forecasting a forecasted action comprises automatically determining a probability of a future application use given the mission state and the user behavior.

7. The method of claim 1 wherein:

the action comprises the user selecting an application;

the forecasting a forecasted action comprises;

automatically determining a probability of a future task given the mission and the degree of task completion of a current task;

automatically determining a future user behavior given the future task; and

automatically determining the forecasted action from the future task and the future user behavior.

8. A method for power management of a battery powered processor based device, said method comprising:

an application operating in a first application configuration with a first application power mode;

receiving an action of a user from the application;

the action comprising a first and second action;

the action further comprising an application use on a processor based device;

receiving a mission state of a mission;

the mission comprises a plurality of tasks and each task reflecting a degree of task completion;

forecasting a forecasted action from at least one of the user action and the mission state; and

determining a second application configuration with a second application power mode for the application based on the forecasted action.

9. The method of claim 8 wherein:

the first application power mode is one of a plurality of predefined power modes for a first task of the plurality of tasks; and

the second application power mode is one of the plurality of predefined power modes for a second task of the plurality of tasks.

10. The method of claim 9 wherein the plurality of predefined power modes comprises a first, a second and a third power predefined power mode.

11. The method of claim 8 further comprising:

an additional application operating in a first additional application configuration with a first additional application power mode;

receiving a user behavior from the additional application; and

determining a second additional application configuration with a second additional application power mode for the additional application based on the forecasted action.

12. The method of claim 11 further comprising:

the first application power mode, the first additional application power mode, the second application power mode and the second additional application power mode each comprising one of a plurality of predefined power modes; and

defining an optimum power mode from the predefined power modes for the application and an optimum power mode from the plurality of predefined power modes for the additional application.

13. The method of claim 11 further comprising defining an optimum power mode from the plurality of predefined power modes for the application and the optimum power mode from the plurality of predefined power modes for the additional application given a power measure, a task state and the plurality of predefined power modes.

14. A method for power management of a battery powered processor based device, said method comprising:

predefining one or more predefined power mode;

providing access to one or more application wherein each application is configured to operate with a processor based device to support at least one task;

the at least one task representing two or more actions to be performed by a user;

receiving a task state of one of the at least one task;

the task state reflecting a degree of task completion;

defining an optimum power mode for one of the one or more application given a power measure, the task state and the one or more predefined power mode; and

outputting an instruction to modify a power resource allocation for the processor based device wherein the instruction to modify a power resource allocation comprises modifying the one or more predefined power mode to the optimum power mode.

15. The method of claim 14 wherein the one or more predefined power mode comprises a first, a second and a third predefined power mode.

16. The method of claim 14 wherein:

the one or more application comprises a first application and a second application;

each of the first application and the second application are configured to operate in at least two of the one or more predefined power mode; and

the step of defining an optimum power mode for the application further comprises defining an optimum power mode for the first application and an optimum power mode for the second application given the power measure, the task state and the one or more predefined power mode.

17. The method of claim 14 further comprising:

defining a second optimum power mode for the first application and a second optimum power mode for the second application given a second power measure, a second task state and the one or more predefined power mode.

18. The method of claim 14 wherein each predefined power mode defines a power demand and a task efficiency for each task.