Electronic authentication system, device and process
Briefly, example methods, apparatuses, and/or articles of manufacture are disclosed that may be implemented, in whole or in part, using one or more processing devices to facilitate and/or support one or more operations and/or techniques for authenticating an identity of a subject. In particular, some embodiments are directed to techniques for authentication of an identity of a subject as being an identity of a particular unique individual based, at least in part, on involuntary responses by the subject to sensory stimuli.
1. A method comprising:
executing one or more first trained neural networks to obtain one or more context features based, at least in part, on first sensor signals generated by one or more first sensors responsive to a physical environment experienced by a subject;
selecting a response signature from among a plurality of response signatures maintained in a repository of response signatures for a particular unique individual based, at least in part, on an association of at least one of the one or more context features with a repository of context features;
executing one or more second trained neural networks to obtain one or more involuntary response features determined based, at least in part, on second sensor signals generated by one or more second sensors, the second sensor signals being based, at least in part on or responsive to one or more involuntary responses of the subject to the physical environment;
electronically authenticating an identity of the subject as being an identity of the particular unique individual based, at least in part, on the one or more context features and the one or more involuntary response features to provide an authentication result; and
updating one or more of the plurality of response signatures for the particular unique individual maintained in the repository based, at least in part, on the authentication result and the obtained one or more context features.
2. The method of claim 1 , wherein electronically authenticating the identity of the subject as being the identity of the particular unique individual further comprises determining a binary authentication result based, at least in part, on application of a Bayesian detector to the one or more involuntary response features and the one or more context features.
3. The method of claim 1 , wherein executing the one or more first trained neural networks to obtain the one or more context features further comprises:
determining a state of the subject as being agitated, ill, rested or tired, or a combination thereof, based, at least in part, on observations of the subject; and
determining the one or more context features based, at least in part, on the determined state.
4. The method of claim 1 , wherein the one or more context features are further determined based, at least in part, on recognition of one or more objects in one or more images captured by a camera co-located with the subject or recognition of one or more recorded sounds, or a combination thereof.
5. The method of claim 4 , wherein the one or more context features are obtained further based, at least in part, on an association of the recognized one or more objects with one or more objects of significance to the particular unique individual.
6. The method of claim 1 , wherein the second sensor signals are generated by one or more second sensors responsive, at least in part, to detected or measured eye blinking, detected or measured eye movement, detected or measured pupillary response, MEG scan signal, body temperature, blood pressure, brain signals or perspiration, or a combination thereof, of the subject.
7. The method of claim 1 , wherein the second sensor signals are based, at least in part on or responsive to one or more P300 brain signals of the subject or one or more steady-state visual evoked potential (SSVEP) signals of the subject, or a combination thereof.
8. The method of claim 1 , wherein the one or more context features are obtained further based, at least in part, on one or more images captured at a camera or signals generated by one or more first sensors.
9. The method of claim 8 , wherein the one or more first sensors comprise at least one environmental sensor or at least one inertial sensor, or a combination thereof.
10. The method of claim 1 , wherein the one or more context features are further based, at least in part on, or responsive to signals from a camera, one or more environmental sensors, one or more inertial sensors, one or more location determining devices, or a combination thereof.
11. The method of claim 1 , wherein the repository of context features are to be maintained in a non-transitory memory, and wherein the method further comprises updating at least a portion of the repository of context features based, at least in part, on at least one of the one or more context features, at least one of the one or more involuntary response features and authentication of the identity of the subject as being the identity of the particular unique individual.
12. The method of claim 1 , wherein electronically authenticating the identity of the subject as being the identity of the particular unique individual to be performed at a frequency based, at least in part, on a security level.
13. The method of claim 1 , wherein the one or more context features are further based, at least in part, on:
a state of the subject determined by the one or more first trained neural networks.
14. The method of claim 1 , wherein:
one or more features of the one or more second trained neural networks are trained based, at least in part, on observations of responses by the particular unique individual to stimuli over time; and
the method further comprises determining the response signature based, at least in part, on the trained one or more features of the one or more second trained neural networks.
15. The method of claim 14 , wherein the stimuli over time to comprise stimuli to physical environments over time.
16. The method of claim 1 , and further comprising determining the one or more involuntary response features further based, at least in part, on an application of a multi-variable probabilistic model to at a first observed involuntary response by the subject to the physical environment and at least a second observed involuntary response by the subject to the physical environment comprising an observed P300 brain signal, eye movement, eye blinking, heart rate, pupillary response, body temperature or blood pressure, or any combination thereof.
17. The method of claim 1 , and further comprising substituting the physical environment with application of an artificial stimulus to the subject.
18. The method of claim 1 , and further comprising providing a prompt to the subject to enter a password in lieu of authenticating based, at least in part, on the one or more involuntary response features and the response signature.
19. A computing device comprising:
one or more processors to:
execute one or more first trained neural networks to obtain one or more context features based, at least in part, on first sensor signals generated by one or more first sensors responsive to a physical environment to be experienced by a subject;
select a response signature from among a plurality of response signatures maintained in a repository of response signatures to be associated with a particular unique individual based, at least in part, on an association of at least one the one or more context features with a repository of context features;
execute one or more second trained neural networks to obtain one or more involuntary response features to be determined based, at least in part, on second sensor signals to be generated by one or more second sensors, the second sensor signals to be based, at least in part on or responsive to one or more involuntary responses of the subject to the physical environment;
authenticate an identity of the subject to be an identity of the particular unique individual based, at least in part, on the one or more context features and the one or more involuntary response features to provide an authentication result; and
update one or more of the plurality of response signatures for the particular unique individual maintained in the repository based, at least in part, on the authentication result and the obtained one or more context features.