Informatisches Kolloquium und CINACS-Kolloquium
Wintersemester 2010/2011

Montag, 15. November 2010
um 17 Uhr c.t.
Vogt-Kölln-Straße 30
Konrad-Zuse-Hörsaal
Gebäude B

Prof. Steffen Werner, Ph.D.
Department of Psychology and Communication Studies
Human Factors Program
University of Idaho

The human element in computer security - Graphical passcodes as a means to create secure authentication systems

User authentication is a central component within the field of HCI security and its importance is growing with the increasing number of electronically available assets such as bank accounts, retirement accounts, medical records, and other personal information. Dwarfing the use of token and biometrically based authentication schemes, authentication via passwords is still the most widespread mechanism used. Common alphanumeric passwords are an example of the larger set of general cognitive authentication methods, which presents a unique optimization problem to cognitive psychologists. The design of a successful "password" system thus can be viewed as the optimization of the "true" entropy of a password space and the ability of a human user to acquire, remember, and reproduce a "password" for an authentication challenge.

In our lab we have investigated a number of alternatives to common alphanumeric passwords - mainly graphical or locimetric passcode systems. Our novel Composite Scene Authentication (CSA) approach assigns users composite visual scenes to increase the memorability of the pseudo-randomly generated passcodes up to a complexity of 46.5 bits.

Initial results in a series of large evaluation studies using more than 600 participants show a clear advantage in memorability of well-designed graphical passcodes over alphanumeric passwords of similar complexity. Not surprisingly, many other popular approaches to graphical passwords fall far short of expectations. Based on our empirical results, central features of a successful graphical authentication system will be highlighted.

Kontakt

Prof. Dr. Christopher Habel

Telefon 2417