Computer Science · Artificial Intelligence
Stronger Hash Functions Through Block Shuffling
Elizabeth Reid, Chris Wilkens, Christina Wright
See on the mapAbstract
In this paper we present a new block chaining approach for itera-tive hash functions based on deterministic block shuffling. We prove that our new scheme does not weaken collision resistance and that it is provably secure against message extension attacks under very spe-cific circumstances. Along the way, we prove that no hash function with a finite state size can be provably secure against message exten-sion attacks if it only reads the input once. Finally, we implement our chaining approach and compare its performance to existing hash functions. 1