Latest Research

Home>Latest Research>HPMA-NTRU: High-Performance Polynomial Multiplication Accelerator for NTRU

December 6, 2022

Return to Latest Research

HPMA-NTRU: High-Performance Polynomial Multiplication Accelerator for NTRU

He P, Tu Y, Khalid A, O’Neill M and Xie J. “HPMA-NTRU: High-Performance Polynomial Multiplication Accelerator for NTRU,” 2022 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFT), 2022, pp. 1-6, DOI: 10.1109/DFT56152.2022.9962336.

Along the rapid development of large-scale quantum computers, post-quantum cryptography (PQC) has drawn significant attention from research community recently as it is proven that the existing public-key cryptosystems are vulnerable to the quantum attacks. Meanwhile, the recent trend in the PQC field has gradually switched to the hardware acceleration aspect. Following this trend, this work presents a novel implementation of a High-performance Polynomial Multiplication hardware Accelerator for NTRU (HPMA-NTRU) under different parameter settings, one of the lattice-based PQC algorithm that is currently under the consideration by the National Institute of Standards and Technology (NIST) PQC standardization process. In total, we have carried out three layers of efforts to obtain the proposed work. First of all, we have proposed a new schoolbook algorithm based strategy to derive the desired polynomial multiplication algorithm for NTRU. Then, we have mapped the algorithm to build a high-performance polynomial multiplication hardware accelerator and have extended this hardware accelerator to different parameter settings with proper adjustment. Finally, through a series of complexity analysis and implementation based comparison, we have shown that the proposed hardware accelerator obtains better area-time complexities than the state-of-the-art one. The outcome of this work is important and will impact the ongoing NIST PQC standardization process and can be deployed further to construct efficient NTRU cryptoprocessors.