Kiran S. Balagani

Matteo Cardaioli, Mauro Conti, Kiran Balagani et al.

Personal Identification Numbers (PIN) are widely used as authentication method for systems such as Automated Teller Machines (ATMs) and Point of Sale (PoS). Input devices (PIN pads) usually give the user a feedback sound when a key is pressed. In this paper, we propose an attack based on the extraction of inter-keystroke timing from the feedback sound when users type their PINs. Our attack is able to reach an accuracy of 98% with a mean error of 0.13 +/-6.66 milliseconds. We demonstrate that inter-keystroke timing significantly improves the guessing probability of certain subsets of PINs. We believe this represents a security problem that has to be taken into account for secure PIN generation. Furthermore, we identified several attack scenarios where the adversary can exploit inter-keystroke timing and additional information about the user or the PIN, such as typing behavior. Our results show that combining the inter-keystroke timing with other information drastically reduces attempts to guess a PIN, outperforming random guessing. With our attack, we are able to guess 72% of the 4-digit PINs within 3 attempts. We believe this poses a serious security problem for systems that use PIN-based authentication.

Jaroslav Sedenka, Kiran Balagani, Vir Phoha et al.

Biometric key generation techniques are used to reliably generate cryptographic material from biometric signals. Existing constructions require users to perform a particular activity (e.g., type or say a password, or provide a handwritten signature), and are therefore not suitable for generating keys continuously. In this paper we present a new technique for biometric key generation from free-text keystroke dynamics. This is the first technique suitable for continuous key generation. Our approach is based on a scaled parity code for key generation (and subsequent key reconstruction), and can be augmented with the use of population data to improve security and reduce key reconstruction error. In particular, we rely on linear discriminant analysis (LDA) to obtain a better representation of discriminable biometric signals. To update the LDA matrix without disclosing user's biometric information, we design a provably secure privacy-preserving protocol (PP-LDA) based on homomorphic encryption. Our biometric key generation with PP-LDA was evaluated on a dataset of 486 users. We report equal error rate around 5% when using LDA, and below 7% without LDA.