Biometric re-authentication: an approach towards achieving transparency in user authentication

Abstract

Providing fixed re-authentication attempts to the user in case of password mismatch is a very old concept. But as password based authentication mechanisms are prone to many security attacks, user’s biometric properties along with passwords are highly in use for user authentication these days. It will be more secure and convenient if re-authentication influenced by the biometric behavior of the user is applied in the authentication system. It is a challenging task to identify a genuine user using behavioral biometric due to its low repeatability and wide variation. So, to increase the efficiency and robustness of the authentication system in case of score mismatch, the clustering of user’s behavior and assigning different re-authentication attempt to different cluster is needed. In this paper, we have proposed a transparent fixed text, keystroke based user authentication framework, which will enhance the security of traditional password based authentication mechanism. A new classification algorithm and dynamic attempt allocation algorithm have been proposed which will make the authentication system smart enough to provide the genuine user a fare authentication attempt.