Multi-modal behavioural biometric authentication for mobile devices
Faculty of Computing, Health and Science
ECU Security Research Institute
The potential advantages of behavioural biometrics are that they can be utilised in a transparent (non-intrusive) and continuous authentication system. However, individual biometric techniques are not suited to all users and scenarios. One way to increase the reliability of transparent and continuous authentication systems is create a multi-modal behavioural biometric authentication system. This research investigated three behavioural biometric techniques based on SMS texting activities and messages, looking to apply these techniques as a multi-modal biometric authentication method for mobile devices. The results showed that behaviour profiling, keystroke dynamics and linguistic profiling can be used to discriminate users with overall error rates 20%, 20% and 22% respectively. To study the feasibility of multi-modal behaviour biometric authentication system, matching-level fusion methods were applied. Two fusion methods were utilised: simple sum and weight average. The results showed clearly that matching-level fusion can improve the classification performance with an overall EER 8%.