Multimodal authentication

• US 8,079,079 B2
• Filed: 06/29/2005
• Issued: 12/13/2011
• Est. Priority Date: 06/29/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Zhang, Zhengyou, Williams, David W., Kong, Yuan, Liu, Zicheng, Kurlander, David, Sinclair, Mike
• Primary Examiner(s): Abrishamkar, Kaveh
• Assistant Examiner(s): Khoshnoodi, Nadia
• Application Number:
  US11/171,145
• Publication Number:
  US 20070005988A1
• Time in Patent Office: 2,358 Days
• Field of Search: None
• US Class Current: 726/19
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06V 10/811 the classifiers operating on different input data, e.g. multi-modal recognition
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: H04L 63/08 for authentication of entities cryptographic mechanisms or cryptographic arrangements for entity authentication H04L9/32
• 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 multimodal system that employs a plurality of sensing modalities which can be processed concurrently to increase confidence in connection with authentication. The multimodal system and/or set of various devices can provide several points of information entry in connection with authentication. Authentication can be improved, for example, by combining face recognition, biometrics, speech recognition, handwriting recognition, gait recognition, retina scan, thumb/hand prints, or subsets thereof. Additionally, portable multimodal devices (e.g., a smartphone) can be used as credit cards, and authentication in connection with such use can mitigate unauthorized transactions.

Claims

1. A system that facilitates authentication, the system comprising:

a processor;

memory communicatively coupled to the processor, the memory having stored therein computer-executable instructions to implement the system, including;

a sensing component that concurrently receives a plurality of sensed inputs and outputs sensing data, wherein the sensed inputs include a current context of data access and at least one authentication input for authentication;

an artificial intelligence component that uses a pattern matching classifier to select the use of a plurality of authentication input subsystems based on the current context, and receives user preference for using at least one of the plurality of authentication input subsystems for authentication, wherein the artificial intelligence component selects the use of the at least one different authentication input subsystem for authentication by employing the pattern matching classifier to perform at least one of probabilistic or statistical-based inference analysis of one or more contexts sensed by the sensing component that are different from the current context; and

an authentication component that performs an authentication process employing the sensing data, and determines whether to enable access to a full set of device features or a subset of the device features based on the current context of data access when the sensing data has resulted in successful authentication.

2. The system of claim 1, wherein the pattern matching classifier includes a support vector machine (SVM) classifier or a naï

ve Bayes classifier.

3. The system of claim 1, wherein the sensing component outputs at least one of speech recognition related information, face recognition related information, handwriting recognition related information, or gait recognition related information.

4. The system of claim 1, wherein the sensing component outputs at least one of a retinal scan related information or handprint related information.

5. The system of claim 1, wherein the sensing component outputs biometric related data.

6. The system of claim 1, wherein the authentication component authenticates a user to a portable wireless device.

7. The system of claim 1, wherein the authentication component authenticates a portable wireless device to a remote system.

8. The system of claim 1, wherein the plurality of authentication input subsystems includes two or more of a face recognition component, a speech recognition component, retinal scan component, a handwriting recognition component, a gait recognition component, a handprint recognition component, or a fingerprint recognition component.

9. The system of claim 1, wherein the authentication component provides an indication to the artificial intelligence component that one or more suspect transactions have been detected from the user, the artificial intelligence component based upon the received indication enables use of one or more additional authentication inputs for authentication of the user.

10. A portable wireless device that employs the system of claim 1, wherein the sensed inputs include at least two biometric inputs from a user, at least one of the biometric inputs is detected from a sensor that is integrated into the portable wireless device, at least one of the biometric inputs is detected from an external device that is not physically connected to the portable wireless device, the authentication component employs sensing data associated with the at least two biometric inputs to authenticate the user.

11. A method of authenticating an entity, the method comprising:

employing a processor executing computer executable instructions to perform the following acts;

creating a plurality of user profiles, wherein at least two user profiles are associated with a user, each of the at least two user profiles are associated with one of a plurality of user authentication contexts, each of the at least two user profiles has one or more user selected inputs for authentication based on the authentication context associated with the user profile;

selecting inputs associated with a user profile using a classifier that employs at least one of probabilistic or statistical-based inference analysis of the authentication context;

receiving sensed input data from the automatically selected inputs based on user interaction with the automatically selected inputs;

processing the sensed input data to output data for authentication processing;

comparing the output data with predetermined user authentication data;

granting user access to data when the output data agrees with the predetermined user authentication data to a predetermined percentage and

adjusting a level of authentication processing to a higher level based on detection of one or more suspect transactions from the user, and automatically selecting one or more additional inputs for authentication based on the higher level.

12. The method of claim 11, further comprising transmitting the output data to a remote location for the authentication processing.

13. The method of claim 11, further comprising reducing the user access when the output data does not agree with the user predetermined user authentication data to a predetermined percentage.

14. The method of claim 11, wherein the controlling user access includes one of enabling a device functionality, disabling the device functionally, or reducing the device functionality.

15. The method of claim 11, wherein the level of authentication processing is a higher level that employs additional authentication inputs for the authentication processing than the automatically selected inputs associated with the user profile.

16. The method of claim 11, wherein the level of authentication processing is a higher level that employs a more reliable authentication input for the authentication processing than the automatically selected inputs associated with the user profile.

17. The method of claim 11, further comprising, upon the user choosing one or more inputs in the authentication context that are changed from the user selected inputs associated with the authentication context, associating the one or more changed user chosen inputs with the user profile associated with the authentication context.

18. The method of claim 11, further comprising performing at least one of the following as one of the selected inputs:

face recognition, speech recognition, gait recognition, handwriting recognition, retinal scan recognition, biometric recognition, or handprint recognition.

19. A system that facilitates authentication of an entity, comprising:

a processor;

memory communicatively coupled to the processor, the memory having stored therein computer-executable instructions performing acts comprising;

sensing a plurality of authentication input subsystems for user authentication;

identifying a current context associated with the authentication, the current context including a type of stored data protected by the authentication;

determining a plurality of distinct biometric inputs required for authentication in the current context;

identifying at least two authentication input subsystems from the plurality of authentication input subsystems that in combination have capability to receive the plurality of the distinct biometric inputs, the at least two authentication input subsystems being identified by a pattern matching classifier that performs at least one of probabilistic or statistical-based inference analysis of the current context associated with the authentication;

receiving the plurality of distinctive biometric inputs via the at least two authentication input subsystems;

processing the sensed multiple distinct biometric inputs concurrently and outputting authentication data;

comparing the authentication data with predetermined authentication data and generating authentication results;

controlling access based on the authentication results; and

requesting one or more additional requested biometric inputs on a periodic basis when the plurality of distinctive biometric inputs results in excessive transactions during permitted access.

20. The system of claim 19, wherein the biometric include physical characteristics of a user as measured and processed into the authentication data, at least one of the physical characteristics includes a facial feature, a voice characteristic, a finger feature, a hand feature, or an eye feature.