Objectives: Behavioural biometric systems capture and use a time-based assessment of how a subject performed a particular task. A series of modalities for mobile device behavioural interaction involve the use of the touch screen – these modalities include signature and doodle production (with a stylus or finger), virtual keyboard keystroke and device swipe. Conventionally when establishing the performance of biometric systems, we assess an overall accuracy measure. It is, however, possible to assess how humans interact with sensors and implementations, and attribute errors to either the technology or the interaction itself. This outcome of this analysis can inform the design of systems, user interfaces, sensor positioning and flow of information provided to the subject. This three-year project will explore mobile touch-screen behavioural biometric performance assessment based on user interaction. Examining a range of modalities as highlighted above for assessing identity, our work will a) explore common interaction errors within a series of donation scenarios b) define a framework for interaction error mitigation and c) define best practice for sample collection scenarios and configuration. This work will provide clear guidance for commercial systems using signature/swipe as a commercial form of identity, thereby enhancing the inclusiveness and robustness of these technologies across the general public on mobile platforms.

Expected Results: There are three expected outcomes from this project: a) a deeper understanding as to how users interact with mobile devices within the touch-screen modalities explored and how the user interface and instructions modulates this interaction in terms of system vs. user performance, b) a framework for mitigation encapsulating best-practice for implementation and design and c) an assessment of touch-screen modality performance across devices and systems in collaboration with our industrial partner. We also anticipate that the outcome of this trial will lead to the adoption of systems design within commercial systems.

Planned secondment(s): Two secondments would take place within this project. The first is a five month visit to UC3M where the ESR will use specialist user monitoring equipment to assess the effects on mobile interaction with respect to environmental and posture conditions and to assess interaction when applied to a wider range of devices and instructional languages. The expected results of this secondment are an understanding of how various environmental conditions and multi-language instructions can be accounted for when assessing the interaction effects of mobile biometric platforms. The second five-month secondment would be with Namirial wherein the ESR would trial a series of user interfaces on a range of commercial devices and software, assessing performance and interaction. The expected outcome of this secondment are a commercially-focused set of results showing how interaction assessment can be used to enhance the design of on-screen interfaces and assistance, and improve the resilience of touch-screen mobile systems to interaction (and device) variability.

Start Date: July 2017