Keystroke dynamics on a mobile handset: a feasibility study

The ability of third generation telephones to store sensitive information, such as financial records, digital certificates and company records, makes them desirable targets for impostors. This paper details the feasibility of a non‐intrusive subscriber authentication technique – the use of keystroke dynamics. This feasibility study comprises a number of investigations into the ability of neural networks to authenticate users successfully based on their interactions with a mobile phone keypad. The initial results are promising with network classification performing well, achieving a 9.8 per cent false rejection rate and an 11.0 per cent false acceptance rate.