Abstract

Elliptic curve cryptosystems proved to be well suited for securing systems with constrained resources like embedded and portable devices. In a fault attack, errors are induced during the computation of a cryptographic primitive, and the faulty results are collected to derive information about the secret key stored into the device in a non-readable way. Scenarios where the secure devices are seized by an opponent are quite common. Consequently, it is possible for an attacker to induce changes in the working environment of the device to cause alterations in the computation of the cryptographic primitive. We introduce a new fault model and attack methodology to recover the secret key employed in implementations of the Elliptic Curve Digital Signature Algorithm. Our attack exploits the information leakage induced when altering the execution of the modular arithmetic operations used in the signature primitive and does not rely on the properties of the underlying elliptic curve mathematical structure, thus being applicable to curves defined on both prime fields and binary fields. The attack is easily reproducible with low cost fault injection technologies relying on transient errors placed within a single datapath width of the target architecture.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call