Performance Analysis for a Touch Dynamic Authentication System with Reduced Feature Set Using Neural Networks

ABSTRACT

Increase in the usage of smart phones increases the burden for the users to memorize many passwords. It has also increased the need for stronger or enhanced authentication mechanisms. Littlemore innovation labs are providing paperless solution for conducting examinations using touch pad that is believed to be a promising technology in future. This paper explores the suitability of using touch dynamics as an additional level of security during authentication. Both security and performance concerns are investigated. A security application developed can be used on a touch screen device capable of imitating someone's typing characteristics. When the user starts interacting with the device, the developed application starts capturing the behavioural features given by the user's swipe. Thirty different features are captured, from which, most prominent features are identified by depending on the usefulness of the feature. Good results with the reduced feature set are obtained, thereby improving the performance of the system using neural network techniques. The results show that touch dynamics on a smart phone are more durable against certain attacks on a personal computer.

Keywords:

• Authentication
• biometrics
• touch dynamics
• security

Acknowledgments

We thank UGC for providing us necessary support for carrying out this project work.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work has been funded by University Grants Commission (UGC) under Minor project, New Delhi [Grant number: No.F MRP-4985/14 (SERO/UGC) Comcode: TNBA025 March 2014].

Notes on contributors

R. SenthilPrabha

R. SenthilPrabha received the Bachelor's degree in electronics and communication engineering and the Master's degree in computer science from Bharathiyar University, Coimbatore. She is currently working as an assistant professor in the Department of Information Technology in PSG College of Technology. Her areas of interest include biometrics and computer networks.

E-mail: [email protected]

R. Vidhyapriya

R. Vidhyapriya received the Bachelor's degree in electrical and electronics engineering and the Master's degree in applied electronics from PSG College of Technology, Coimbatore. The doctoral degree was awarded in the year 2008 by Anna University, Chennai. She is currently working as a professor in the Department of Information Technology in PSG College of Technology. Her areas of interest include mobile communication, computer communication networks, network management systems, and digital design.

E-mail: [email protected]

N. RavithaRajalakshmi

N. RavithaRajalakshmi received the Bachelor's degree in information technology and the Master’s degree in biometrics and cyber security at PSG College of Technology, Coimbatore. She is awarded the Best Out Going Student by the Department of Information Technology and currently working as an assistant professor in the Department of IT, PSG College of Technology.

E-mail: [email protected]