Radar-based behaviometric user authentication
A first behaviometric user profile for a first user is generated and stored, by detecting a position and velocity of the first user relative to the mobile device based on a received response from a radar transmission while the first user uses the mobile device, the received response over time indicating a position and velocity of the first user. Based on further received responses of additional radar transmissions an additional behavioral pattern of an unknown user is determined. The additional behavioral pattern is then compared to the first behaviometric user profile, and based on the comparison, a measure of similarity between the first behaviometric user profile and the additional behavioral pattern, measuring if the first user and the unknown user are a same user is heuristically determined. As a result of the comparison, operation or access to at least some data stored on the mobile device is prevented.
1. A method of authenticating a user using a mobile device, the method comprising:
generating and storing a first behaviometric user pattern for an authenticated user, said pattern including a position and velocity of said authenticated user relative to said mobile device, said position and velocity of said authenticated user relative to said mobile device determined by a received response from a radar transmission sent by said mobile device while said first user uses said mobile device, said radar transmission transmitted piece-wise continuously over time such that said received response indicates a position and velocity of said authenticated user over time;
determining, based on further received responses of additional radar transmissions sent by said mobile device, an additional behavioral pattern of an unknown user, said additional behavioral pattern including a position and velocity of said unknown user relative to said mobile device, said radar transmission being sent after a first authentication of said authenticated user by way of another radar transmission;
comparing said additional behavioral pattern to said first behaviometric user pattern;
determining heuristically, based on said comparing, a measure of similarity between said first behaviometric user pattern and said additional behavioral pattern;
determining, based on said measure of similarity, if said unknown user is said authenticated user;
preventing operation or access to at least some data stored on said mobile device if said unknown user is not said authenticated user, as determined by said comparison of said additional behavioral pattern to said first behaviometric user pattern; and
granting operation or access to at least some data stored on said mobile device if said unknown user is said authenticated user, as determined by said comparison of said additional behavioral pattern to said first behaviometric user pattern;
wherein said determining said additional behavioral pattern is synchronized in time with receipt, from said unknown user, of user responses to user prompts.
2. The method of claim 1 , wherein said unknown user is said authenticated user, said method further comprising, prior to said determining said additional behavioral pattern, authenticating said authenticated user on said mobile device by detecting biometric or behavioral characteristics related to a gesture or movement of said authenticated user or said mobile device.
3. The method of claim 2 , wherein said additional behavioral pattern reinforces said authentication of said authenticated user based on said biometric or behavioral characteristics.
4. The method of claim 1 , wherein said received response for purpose of authentication from said radar transmission comprises a received response from a LIDAR (LIght Detection And Ranging) radar transmission.
5. The method of claim 1 , wherein said received response from said radar transmission comprises a received response from at least two radar transmissions of different types.
6. The method of claim 1 , wherein said user prompts are implicitly provided, are hidden from said unknown user, and are based on said unknown user's desire to reach their goal without perceivable interference.
7. The method of claim 6 , wherein said implicitly provided user prompts include at least one of an open pop-up advertising banner prompting the user to close said pop-up advertising banner, a beeping or music sound prompting the user to terminate said beeping or music sound, and a small window prompting the user to resize the window.
8. The method of claim 1 , further comprising, prior to said generating said authenticated user pattern and to said determining said additional behavioral pattern, determining times or types of movements to be included in said authenticated user pattern and said additional behavioral pattern.
9. The method of claim 1 , further comprising, using a machine learning based classifier to distinguish between said additional behavioral patterns stemming from different users.
10. The method of claim 1 , wherein said detected position and velocity of said authenticated user relative to said mobile device comprise a detected position and velocity of a finger or a hand of said authenticated user relative to said mobile device.