A multimodal biometric identification method for mobile applications security

Abstract

This paper presents a design approach of a reliable authentication system for mobile applications (such as those within m-Health or m-Banking areas). This means that the biometric data processing should optimize the security performance vs. the computational complexity. The security is given by the combination of fingerprint, iris and voice features that define the multimodal pattern of an individual. The complexity reduction is supported by a reduced feature space, especially for the fingerprint and iris recognition components of the overall system.