Authentication involving authentication operations which cross reference authentication factors

• US 8,925,058 B1
• Filed: 03/29/2012
• Issued: 12/30/2014
• Est. Priority Date: 03/29/2012
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Dotan, Yedidya, Duane, William M., Linn, John, Hodgman, Roy, Lin, Derek
• Primary Examiner(s): Edwards, Linglan
• Assistant Examiner(s): GRACIA, GARY S
• Application Number:
  US13/434,257
• Time in Patent Office: 1,006 Days
• Field of Search: 726/7
• US Class Current: 726/7
• CPC Class Codes: G06F 21/31 User authentication
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: H04L 63/0861 using biometrical features, e.g. fingerprint, retina-scan cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231

Abstract

A technique of authenticating a person involves obtaining, during a current authentication session to authenticate the person, a first authentication factor from the person and a second authentication factor from the person, at least one of the first and second authentication factors being a biometric input. The technique further involves performing an authentication operation which cross references the first authentication factor with the second authentication factor. The technique further involves outputting, as a result of the authentication operation, an authentication result signal indicating whether the authentication operation has determined the person in the current authentication session likely to be legitimate or an imposter. Such authentication, which cross references authentication factors to leverage off of their interdependency, provides stronger authentication than conventional naïve authentication.

Claims

1. In electronic circuitry, a method of authenticating a person, the method comprising:

during a current authentication session to authenticate the person, obtaining a first authentication factor from the person and a second authentication factor from the person, at least one of the first and second authentication factors being a biometric input;

performing an authentication operation which cross references the first authentication factor with the second authentication factor; and

outputting, as a result of the authentication operation, an authentication result signal indicating whether the authentication operation has determined the person in the current authentication session likely to be legitimate or an imposter,

wherein performing the authentication operation which cross references the first authentication factor with the second authentication factor includes;

generating a correlation result based on an amount of correlation between the first authentication factor and the second authentication factor;

wherein the authentication result signal is based on the correlation result;

wherein the first authentication factor is a first biometric input;

wherein the second authentication factor is a second biometric input; and

wherein generating the correlation result based on the amount of correlation between the first authentication factor and the second authentication factor includes providing, as the correlation result, a correlation score indicating the amount of correlation between the first biometric input and the second biometric input;

wherein performing the authentication operation further includes providing the correlation result to a risk engine which is constructed and arranged to perform an adaptive authentication operation based on the correlation result, the authentication result signal being outputted by the risk engine and being based on the correlation result;

further comprising the risk engine comparing authentication factors from the current authentication session to expected authentication factors, wherein the expected authentication factors are received from a user profile; and

further comprising performing a vitality test, adjusting the test for circadian rhythm, age and elapsed time since a last successful authentication, and comparing against the user profile.

2. A method as in claim 1 wherein performing the authentication operation further includes:

performing multifactor authentication based on (i) the first authentication factor, (ii) the second authentication factor, and (iii) the correlation result to produce the authentication result signal.

3. A method as in claim 2 wherein performing multifactor authentication includes:

performing a first comparison operation which compares the first authentication factor to a first expected factor,

performing a second comparison operation which compares the second authentication factor to a second expected factor, and

performing an evaluation of the correlation result; and

wherein the authentication result signal is based on results of the first comparison operation, the second comparison operation and the evaluation of the correlation result.

4. A method as in claim 3 wherein the correlation result is a correlation score indicating the amount of correlation between the first authentication factor and the second authentication factor;

and wherein performing the evaluation of the correlation result includes;

performing a third comparison operation which compares the correlation score to a predefined correlation threshold, the authentication result signal being based on results of the first comparison operation, the second comparison operation and the third comparison operation.

5. A method as in claim 3 wherein outputting the authentication result signal indicating whether the authentication operation has determined the person in the current authentication session likely to be legitimate or an imposter includes:

indicating (i) successful authentication only when all of the comparison operations provide positive outcomes, and (ii) unsuccessful authentication when at least one comparison operation does not provide a positive outcome.

6. A method as in claim 1 wherein the correlation result is a correlation score indicating the amount of correlation between the first authentication factor and the second authentication factor;

and wherein providing the correlation result to the risk engine includes;

inputting, as a risk-based authentication factor, the correlation score into the risk engine, the authentication result signal being based on the correlation score.

7. A method as in claim 6 wherein performing the authentication operation further includes:

inputting, as additional risk-based authentication factors, the first and second authentication factors into the risk engine, the authentication result signal being further based on the first and second authentication factors.

8. A method as in claim 1 wherein the first authentication factor is a biometric input;

wherein the second authentication factor is a non-biometric input; and

wherein generating the correlation result based on the amount of correlation between the first authentication factor and the second authentication factor includes providing, as the correlation result, a correlation score indicating the amount of correlation between the biometric input and the non-biometric input.

9. A method as in claim 8 wherein the biometric input includes a current set of facial images of the person captured during the current authentication session through a digital camera of a computerized device;

wherein the non-biometric input includes a current time of the current authentication session; and

wherein providing the correlation score includes (i) generating an expected measurement of the current set of facial images of the person based on a previous measurement of a previous set of facial images of the person captured during a previous authentication session, and (ii) comparing a current measurement of the set of facial images to the expected measurement of the set of facial images.

10. A method as in claim 9 wherein performing the authentication operation further includes:

comparing the current set of facial images of the person against a previously generated template.

11. A method as in claim 9 wherein the current measurement of the current set of facial images includes a current pulse waveform derived from the current set of facial images;

wherein the expected measurement of the current set of facial images includes an expected pulse waveform for the current set of facial images; and

wherein providing the correlation score further includes generating a difference between the current pulse waveform and the expected pulse waveform.

12. A method as in claim 1 wherein the biometric input includes current video of the person captured during the current authentication session through a digital camera;

wherein the second biometric input includes current audio of the person captured during the current authentication session through a microphone;

wherein the digital camera and the microphone are co-located on a common computing device; and

wherein providing the correlation score includes (i) comparing the current video with the current audio and (ii) generating, as the correlation score, a value indicating how well the current video matches with the current audio.

13. A method as in claim 12 wherein performing the authentication operation which cross references the first authentication factor with the second authentication factor further includes:

performing a dynamic facial recognition operation based on the current video of the person captured during the current authentication session through the digital camera, and

performing a dynamic voice recognition operation based on the current audio of the person captured during the current authentication session through the microphone.

14. A method as in claim 1 wherein one of the first and second biometric input includes video input which provides cardiac identification information about the person.

15. A method as in claim 14 wherein the cardiac information identifies a pulse waveform for use in (i) identifying the person and (ii) indicating vitality as part of a concurrent vitality test.

16. A method as in claim 1 wherein performing the authentication operation which cross references the first authentication factor with the second authentication factor includes generating a correlation result based upon an interdependence between the first authentication factor and the second authentication factor.

17. A method as in claim 16 wherein generating the correlation result based upon the interdependence further includes at least one of applying fuzzy logic algorithms, generating weight values for each authentication factor, and applying neural nets, to generate the correlation result for the cross referenced authentication factors.

18. A method as in claim 1, wherein:

performing the authentication operation includes a server receiving the first and second authentication factors during the current authentication session and generating the correlation result;

wherein providing the correlation score includes providing the correlation score to a risk engine; and

wherein outputting an authentication result signal includes generating a risk-based score based on the correlation result and the authentication factors is performed by the risk engine.

19. A method as in claim 1, further comprising generating, by the risk engine, a risk score based upon the vitality test, comparing the risk score to a risk threshold, and outputting an authentication result signal.

20. An electronic system to authenticate a person, the electronic system comprising:

receiving circuitry to obtain, during a current authentication session to authenticate the person, a first authentication factor from the person and a second authentication factor from the person, at least one of the first and second authentication factors being a biometric input;

processing circuitry coupled to the receiving circuitry to perform an authentication operation which cross references the first authentication factor with the second authentication factor; and

outputting circuitry coupled to the processing circuitry to output, as a result of the authentication operation, an authentication result signal indicating whether the authentication operation has determined the person in the current authentication session likely to be legitimate or an imposter;

wherein the processing circuitry performing the authentication operation which cross references the first authentication factor with the second authentication factor is constructed and arranged to generate a correlation result based on an amount of correlation between the first authentication factor and the second authentication factor;

wherein the authentication result signal is based on the correlation result;

wherein the first authentication factor is a first biometric input;

wherein the second authentication factor is a second biometric input; and

wherein, when constructed and arranged to generate the correlation result based on the amount of correlation between the first authentication factor and the second authentication factor, the processing circuitry is further constructed and arranged to provide, as the correlation result, a correlation score indicating the amount of correlation between the first biometric input and the second biometric input;

wherein the processing circuitry performing the authentication operation further includes providing the correlation result to a risk engine which is constructed and arranged to perform an adaptive authentication operation based on the correlation result, the authentication result signal being outputted by the risk engine and being based on the correlation result;

further comprising the risk engine comparing authentication factors from the current authentication session to expected authentication factors, wherein the expected authentication factors are received from a user profile; and

further comprising performing a vitality test, adjusting the test for circadian rhythm, age and elapsed time since a last successful authentication, and comparing against the user profile.

21. A computer program product which includes a non-transitory computer readable medium storing a set of instructions to authenticate a person, the set of instructions causing computing circuitry to perform a method, comprising:

during a current authentication session to authenticate the person, obtaining a first authentication factor from the person and a second authentication factor from the person, at least one of the first and second authentication factors being a biometric input;

performing an authentication operation which cross references the first authentication factor with the second authentication factor; and

outputting, as a result of the authentication operation, an authentication result signal indicating whether the authentication operation has determined the person in the current authentication session likely to be legitimate or an imposter;

wherein performing the authentication operation which cross references the first authentication factor with the second authentication factor includes;

generating a correlation result based on an amount of correlation between the first authentication factor and the second authentication factor;

wherein the authentication result signal is based on the correlation result;

wherein the first authentication factor is a first biometric input;

wherein the second authentication factor is a second biometric input; and

wherein generating the correlation result based on the amount of correlation between the first authentication factor and the second authentication factor includes providing, as the correlation result, a correlation score indicating the amount of correlation between the first biometric input and the second biometric input;

wherein performing the authentication operation further includes providing the correlation result to a risk engine which is constructed and arranged to perform an adaptive authentication operation based on the correlation result, the authentication result signal being outputted by the risk engine and being based on the correlation result;

further comprising the risk engine comparing authentication factors from the current authentication session to expected authentication factors, wherein the expected authentication factors are received from a user profile; and

further comprising performing a vitality test, adjusting the test for circadian rhythm, age and elapsed time since a last successful authentication, and comparing against the user profile.