Contribution of TEE and Parallel Computing to Performance and Security of Biometric Authentication Improvement

Abstract

In this work, we present a new idea for improving biometric authentication of personal identities on mobile devices with significant accuracy and convenience because the problems that arise here are the vulnerability and confidentiality of biometric data before, during and after a biometric recognition operation, and the low processing speed during the same operation. The aim of this work is to propose a combination of the concepts of parallel computing and Trusted Execution Environments (TEE) as a solution to the problems raised. We thus address how parallel computing can be useful in accelerating the recognition of individuals, by reducing the interaction time with the database thus preventing some vulnerabilities. Finally, a hardware-assisted technology namely Intel SGX (Software Guard Extensions) is proposed for practical implementations.