Researchers at Jeppiaar Engineering College in Chennai, India, have developed a biometric keystroke algorithm that learns how the user types, which they say could serve as a better authentication method than entering usernames and passwords.
The dynamic characteristics of an individuals typing or the human typing pattern varies from person to person, and could be used to identify people online, according to the researchers. In addition, they say their keystroke method of authentication could be less expensive and more reliable than conventional systems.
"To find out the typing patterns we do not require any additional software," the researchers note.
They say the algorithm uses various software systems that are already present in computers, which leads to decreased costs. The system also relies on new user registration where the keystrokes are recorded, as well as the authentication of the stored keystroke data.
The researchers tested their method on five samples with a corresponding user name and password.
Other researchers are developing voice recognition biometric systems and emoji-based passcodes. Analysts say it could be easier for users to remember emoji-based passcodes compared to traditional passwords.
From Network World
View Full Article
Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA