Improving graphical password resistant to shoulder-surfing using 4-way recognition-based sequence reproduction (RBSR4)

Abstract

Passwords are widely used when accessing computers, networks, accounts and websites. A big draw back of password is called password problem which is not being resist against several password attacks such as guessing, dictionary attack, key-loggers, shoulder-surfing and social engineering. Other than being secure against these attacks passwords should be easy to remember. Usability and security are two important issues to be concerned about while working with passwords. Graphical passwords seem to be the solution as it is described more in the paper. A graphical password is an authentication system that works by having the user select from images, in a specific order, presented in a graphical user interface (GUI). Despite the high standards of Graphical Passwords, they are still vulnerable to some kinds of attacks. Our goal is to propose a new Graphical Password scheme that takes advantage of graphical input displays capabilities to achieve better security than text-based passwords. The proposed research is an approach to enhance the existing Graphical Password techniques and resist against attacks like Shoulder Surfing. This system can be improved to provide a wider password space if more server variables are involved (such as date). Study on robustness of the system against Sniffing can be suggested for further study.