Tools
Tools
Sankar, Siddharth (2025) Continuous User Authentication on Mobile and Desktop Devices using Behavioural Biometrics – a Multimodal Approach. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (937kB) | Preview |
Abstract
Traditional authentication methods are static and it’s the year 2025. There is an exponential growth of mobile and desktop users, the vulnerabilities of these static methods have been exposed. The vulnerability this thesis aims to find a solution for is User’s Session Hijacking. Continuous User Authentication (CUA) using behavioural biometrics can be a solution, but most approaches follow single modality that can struggle with high error rates due to noise and environmental factors, which has limited their widespread adoption. This thesis investigates the effectiveness of a Multimodal CUA framework that will fuse desktop keystroke data with touch gestures from a mobile to create a security layer across the user’s devices.
This thesis adopted a quantitative experimental design, that pairs users from two different datasets to create 41 subjects with a unified identity. The analysis was done by doing a score-level fusion and using the Random Forest classifier to produce an Equal Error Rate.
The Results demonstrated a significant jump in security when fusing the two modalities together instead of using them independently. It was also found that the mobile only modality was insufficient to be used as a standalone authentication system since it achieved error rates of around 20% whereas the fused model achieved scores of around 5% showing a significant reduction of around 75% error rates. This thesis provides a blueprint for a scalable solution to Continuous Authentication using Behavioural Biometrics that enhances security while still maintaining user experience to promote widespread adoption.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Heffernan, Niall UNSPECIFIED |
| Uncontrolled Keywords: | Continuous Authentication; Behavioural Biometrics; Fusion; Multimodal; Random Forest; Session Hijacking; Equal Error Rate; Security |
| Subjects: | T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Biometric Identification Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning |
| Divisions: | School of Computing > Master of Science in Cyber Security |
| Depositing User: | Ciara O'Brien |
| Date Deposited: | 04 Sep 2026 09:41 |
| Last Modified: | 04 Sep 2026 09:41 |
| URI: | https://norma.ncirl.ie/id/eprint/9824 |
Actions (login required)
 |
View Item |