Hiding Identities in Spin Glasses

Biometric traits are increasingly viewed as the ultimate proof of identity. Unfortunately, biometrics also raises many privacy concerns. In order to overcome these, template protection technologies have been developed in which privacy protection is embedded at the technological level. Standard template protection schemes, however either require secret key management or are vulnerable to brute force and cross-matching attacks and they are unsuitable for identification purposes. Here I review a new key-binding mechanism that I have recently proposed and explain why it is expected to be robust against brute force and cross-matching attacks and how it can be used for identification. The security principle underlying this new template protection scheme is the complexity of the energy landscape of spin glasses. The idea is to exploit the mixed ferromagnetic and spin glass phase of the Hopfield neural network to encode the key as a local minimum configuration of the energy functional, "lost" amidst the exponentially growing number of valleys and minima representing the spin glass. The correct fingerprint will be able to retrieve the key by dynamical evolution to the nearest attractor. Other fingerprints will be driven far away from the key.