Techniques for Continuous Touch-Based Authentication

Abstract

The field of continuous touch-based authentication has been rapidly developing over the last decade, creating a fragmented and difficult-to-navigate area for researchers and application developers alike. In this study, we perform a systematic literature analysis of 30 studies on the techniques used for feature extraction, classification, and aggregation in continuous touch-based authentication systems as well as the performance metrics reported by each study. Based on our findings, we design a set of experiments to compare the performance of the most frequently used techniques in the field under clearly defined conditions. In addition, we introduce two new techniques for continuous touch-based authentication: an expanded feature set (consisting of 149 unique features) and a multi-algorithm ensemble-based classifier. The comparison includes 13 feature sets, 11 classifiers, and 5 aggregation methods. In total, 204 model configurations are examined and we show that our novel techniques outperform the current state-of-the-art in each category. The results are also validated across three different publicly available datasets. Our best performing model achieves 4.8% EER using 16 consecutive strokes. Finally, we discuss the findings of our investigation with the aim of making the field more understandable and accessible for researchers and practitioners.