TWC: Small: On the Design of Secure Hash Functions and Block Ciphers

New York University, New York NY

Investigators

Abstract

This project revisits the basic design principle for constructing secure hash functions, block ciphers, and various important cryptographic primitives which are built from them, by investigating new types of constructions that are based on firmer theoretical foundations, and yet are still efficient enough for practical use. In particular, the project focuses on analyzing and improving the use of hash functions and block ciphers as message digests, key derivation functions, message authentication codes, stream ciphers, commitment schemes and random oracles. Additionally, the investigators study novel modes of operation to build complex variable-length primitives from simpler, fixed length components, such as block ciphers and fixed-length compression functions. The project also examines the feasibility of provably immunizing cryptographic algorithms and standards against potentially unknown backdoors. Finally, the project builds firmer foundations for analyzing cryptographic schemes in the idealized security models, such as the random oracle and the ideal cipher models. The project aims to yield more secure hash functions and block ciphers, and more secure and/or efficient usage of hash functions and block ciphers in important cryptographic applications. Besides advancing the theory of cryptography, this project also impacts the real-world design of secure systems, by suggesting more sound use of various cryptographic building blocks for such applications. The PI regularly teaches courses in cryptography and network security, and is actively incorporating the new results into the courses he teaches. In addition, the proposal has a significant graduate student and postdoc training component.