Utilization of keyboard dynamics for unique identification of human users

Abstract

For the past two decades, as the role of computer technology has expanded, computer security (secure processors, information encryption, network protection with biometrics) has also become increasingly important [1-21]. Many computer security schemes have been developed over this time period; however, none of these are foolproof yet. A username and pass-word combination is one of the most popular approaches to securing a computer system, but this system is vulnerable because passwords can be stolen or cracked. However, research suggests that it is possible to identify a user based strictly on their typing style, using pattern recognition, neural networks, and other techniques. This paper focuses on using the keyboard dynamics of the user's password to add an extra layer of security to the natural log-in process.