EASIMask – Towards Efficient, Automated, and Secure Implementation of Masking in Hardware

Das Paper "EASIMask - Towards Efficient, Secure, and Automated Implementation of Masking in Hardware" von Fabian Buschkowski, Pascal Sasdrich und Tim Güneysu wurde bei der DATE'23 akzeptiert.

Abstract:

Side-Channel Analysis (SCA) is a major threat to implementations of mathematically secure cryptographic algorithms. Applying masking countermeasures to hardwarebased implementations is both time-consuming and error-prone due to side-effects buried deeply in the hardware design process.
As a consequence, we propose our novel framework EASIMASK in this work. Our semi-automated framework enables designers that have little experience with hardware implementation or physical security and the application of countermeasures to create a securely masked hardware implementation from an abstract description of a cryptographic algorithm. Its design-flow dismisses the developer from many challenges in the masking process of hardware implementations, while the generated implementations match the efficiency of hand-optimized designs from experienced security engineers. The modular approach can be mapped to arbitrary instantiations using different languages and transformations. We have verified the functionality, security, and efficiency of generated designs for several state of the art symmetric cryptographic algorithms, such as Advanced Encryption Standard (AES), Keccak, and PRESENT.