Stochastic Systems Group
Home Research Group Members Programs  
Demos Calendar Publications Mission Statement Alumni

SSG Seminar Abstract


Secure Biometrics Via Syndromes


Emin Martinian
Mitsubishi Electric Research Laboratory


Biometrics such as fingerprints, irises, and faces are often used for authentication, access control, and encryption instead of passwords. While it is well known that passwords should never be stored in the clear, current systems often store biometrics in the clear and are easily compromised.

I will describe the secure biometric storage problem by discussing the insecurities in current systems, the reasons why password hashing/encryption algorithms are unsuitable for biometrics, and propose a secure biometric storage framework based on low density syndrome codes and the Slepian-Wolf theorem. Finally, I will present experimental results of a prototype for iris biometrics and discuss some of the open problems in this area.




Problems with this site should be emailed to jonesb@mit.edu