Toward a continuous authentication system using a biologically inspired machine learning approach: A case study

Abstract

Smartphones have recently seen massive growth in usage and become a repository for many types of personal information. The privacy and security are primary concerns for their usage, where there is a need to provide seamless and continuous authentication systems(CASs) for smartphones. We introduce in this work a proof-of-concept design and a case study for using a biologically-inspired and hardware-friendly CAS. Our proposed design adopts a hybrid Liquid State Machine (LSM) approach to perform automatic features extraction and multi-modal fusion scheme for users' interaction. Our work establishes the design concepts for future on-chip and explains why LSM can be a promising approach for fast, adaptive and reliable hardware CASs. The experimental part reveals our proof-of-concept testing results against the golden standard features. Furthermore, The results indicate a promising future for such design and a potential biologically-inspired CASs.