Enhancing Side-Channel Attack Resistance of Postquantum Cryptographic Algorithm CRYSTALS-Kyber Using High-Order Chebyshev Filters
Chung‐Wei Kuo, Wei-Chih Hong, Yi-Kai Hsu, Yu-Yi Hong · 2025
As quantum computing progresses, traditional public key cryptographic algorithms are increasingly at risk of compromise, making post-quantum cryptography (PQC) critical for maintaining information security. In August 2024, NIST published its first post-quantum key encapsulation mechanism (ML-KEM), which is based on the CRYSTALS-Kyber algorithm with minor adjustments. Kyber, rooted in lattice-based cryptography, offers efficient encryption and decryption with strong resistance to quantum attacks. However, side-channel attacks (SCAs) employing artificial intelligence techniques pose a significant threat, enabling attackers to extract sensitive information through power trace analysis—even in masked Kyber implementations. To counteract this, we propose the approach of incoporating high-order Chebyshev filter to reduce identifiable features in power traces produced during encryption and weaken SCA models' ability to pinpoint sensitive data. Our results demonstrate that this method reduces the attack success rate from 96% to 61%, providing a substantial defense mechanism for Kyber. This study highlights the potential of high-order Chebyshev filtering as a robust countermeasure against SCA on Kyber, underscoring its importance in strengthening quantumresistant protections.