Developing a biometric- based privacy- aware authentication method for smart devices

Abstract

There are several forms of security can be used to protect the confidentiality and authenticity of data, securely identify rightful owner, or protect user privacy. This thesis analyze the popular authentication systems and propose using biometrics for a more natural and reliable solution. Fuzzy Vault, the key approach for integrating biometric and cryptography, was introduced by Juels and Sudan [7] to overcome biometrics’ lack of template protection and noise handling. Several approaches based on Fuzzy Vault were also introduced in recent research [8-13]. From the flaws of these existing approaches, I proposed new scheme that contains two new module: chaff point generator and chaff point verifier. These module are designed to be integrated easily into the existing systems as well as simple to enhance. To evaluate proposed scheme, I also developed a simulation of authentication system and an Android application. Result from the experiment shows promising improvement for Fuzzy Vault.