Pipeline Oriented Implementation of NORX for ARM Processors

Abstract

NORX is a family of authenticated encryption algorithms that advanced to the third-round of the ongoing CAESAR competition for authenticated encryption schemes. In this work, we investigate the use of pipeline optimizations on ARM platforms to accelerate the execution of NORX. We also provide benchmarks of our implementation using NEON instructions. The results of our implementation show a speed improvement up to 48% compared to the state-of-art implementation on Cortex-A ARMv8 and ARMv7 processors.