Deployment of Keystroke Analysis on a Smartphone

Document Type

Conference Proceeding


Security Research Centre, School of Computer and Security Science, Edith Cowan University


Computing, Health and Science


School of Computer and Security Science, Centre for Security Research




This article was originally published as: Buchoux, A., & Clarke, N.L. (2008). Proceedings of the 6th Australian Information Security Management Conference. Edith Cowan University, Perth, Western Australia.


The current security on mobile devices is often limited to the Personal Identification Number (PIN), a secretknowledge based technique that has historically demonstrated to provide ineffective protection from misuse. Unfortunately, with the increasing capabilities of mobile devices, such as online banking and shopping, the need for more effective protection is imperative. This study proposes the use of two-factor authentication as an enhanced technique for authentication on a Smartphone. Through utilising secret-knowledge and keystroke analysis, it is proposed a stronger more robust mechanism will exist. Whilst keystroke analysis using mobile devices have been proven effective in experimental studies, these studies have only utilised the mobile device for capturing samples rather than the more computationally challenging task of performing the actual authentication. Given the limited processing capabilities of mobile devices, this study focuses upon deploying keystroke analysis to a mobile device utilising numerous pattern classifiers. Given the trade-off with computation versus performance, the results demonstrate that the statistical classifiers are the most effective.