Semester

Fall

Date of Graduation

2007

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Bojan Cukic.

Abstract

Most computer systems use usernames and passwords for authentication and access control. For long, password security has been framed as a tradeoff between user experience and password security. Trading off one for the other appears to be an inevitable dilemma for single password based security applications. As a new biometric for authenticating access, keystroke dynamics offers great promises in hardening the password mechanism. Our research first investigate the keystroke dynamics based password security by conducting an incremental study on user's habituation process for keystroke dynamics analysis using two distinct types of passwords. The study shows that (1) long and complex passwords are more efficient to be employed in keystroke dynamics systems; and (2) there is a habituation and acclimation process before the user obtains a stable keystroke pattern and the system collects enough training data. Then, based on our findings, we propose a two passwords mechanism that attempts to strike the right balance over user experience and password security by adopting a conventional easy-to-memorize password followed by a long-and-complex phrase for keystroke dynamics verification. Analysis and experimental studies successfully demonstrate the effectiveness of our proposed approach.

Share

COinS