User authentication

1. A method, comprising:

authenticating a user of a device using a static authentication technique;

loading a behavior profile associated with the user, where the behavior profile describes patterns of device usage behavior by the user in a three-dimensional space over time slices, and where the behavior profile identifies distinctive user habits;

detecting behavior gathering capabilities of the device;

monitoring usage of the device based, at least in part, on the detected behavior gathering capabilities of the device;

periodically updating a behavior similarity index that describes a similarity between the usage of the device and the patterns of device usage behavior, where the behavior similarity index is weighted based on the distinctive user habits and discounts attributes of the patterns of device usage behavior that the device is incapable of gathering; and

restricting access to the device when the behavior similarity index reaches a predefined threshold.

2. The method of claim 1 , comprising generating the behavior profile.

3. The method of claim 1 , comprising updating the behavior profile.

4. The method of claim 1 , where the patterns of device usage include information describing one or more of, hand motion data, mouse motion data, keystroke metrics, and stress data.

5. The method of claim 4 , where the hand motion data is obtained by tracking motion of a sensory device.

6. The method of claim 1 , where the static authentication technique is one of, image based authentication, password authentication, voice authentication, and possession based authentication.

7. A device, comprising:

a processor;

a set of input components;

an electronic data store in digital communication with the processor, the electronic data store to store behavior profiles associated with a set of users, where the behavior profiles describe patterns of device usage behavior associated with respective users;

a behavior tracking module to gather device usage information as the device is being operated;

a behavior comparison module to periodically generate a behavior similarity index that describes a similarity between the device usage information and a selected behavior profile in the data store,

wherein the behavior comparison module generates the behavior similarity index by selecting a set of features based on:

the set of input components, and

distinctive user habits associated with the selected user profile: and

an authentication module to provide initial access to the device upon successfully authenticating a user based on a credential associated with the user, to designate, based on the successful authentication, the selected behavior profile used by the behavior comparison module, and to restrict access to the device when the behavior similarity index reaches a predefined threshold.

8. The device of claim 7 , comprising a set of input components, where the input components are used by the behavior tracking module to generate the device usage information.

9. The device of claim 8 , where the set of input components include a hand motion tracking component that receives hand motion data during usage of the device.

10. The device of claim 7 , comprising a profile generation module to update a behavior profile based on device usage information gathered by the behavior tracking module.

11. The device of claim 9 , where the behavior comparison module generates the behavior similarity index by applying one of a distance measure and a kernel learning technique to the set of features.

12. A non-transitory computer-readable medium storing computer executable instructions that, when executed, cause the computer to:

obtain a behavior profile associated with a user in response to a successful static authentication by the user to a device, where the behavior profile describes a physical pattern of device usage associated with the user, and where the behavior profile identifies a distinctive user habit;

detect input gathering capabilities of the device;

periodically update a behavior similarity index by comparing physical actions taken during usage of the device to the pattern of device usage associated with the user, where the behavior similarity index takes into account the input gathering capabilities of the device, where the behavior similarity index emphasizes the distinctive user habit when the input gathering capabilities of the device facilitate detecting the distinctive user habit, and where the behavior similarity index is updated by applying one of a distance measure and a kernel learning technique to a set of features associated with the distinctive user habit; and

restrict access to the device when the behavior similarity index reaches a predefined threshold.

13. The non-transitory computer-readable medium of claim 12 , where the instructions further cause the computer to build the behavior profile during a setup period.

14. The non-transitory computer-readable medium of claim 13 where the user is instructed to perform a specific set of tasks when building the behavior profile.