Work

Invited Paper: Enhancing Privacy-Preserving Computing with Optimized CKKS Encryption: A Hardware Acceleration Approach

Year: 2024

Type: conference-paper

Abstract: The widespread adoption of the FHE (Fully Homomorphic Encryption) CKKS (Cheon-Kim-Kim-Song) encryption algorithm is stunted by its slow software implementation, driving the need for efficient hardware acceleration solutions. This paper presents a novel approach to address the challenge by optimizing a segment of the CKKS encryption and decryption process. Using the Residue Number System (RNS), the inputs are encoded into 32-bit representations and processed through the number theoretic transform (NTT). This innovative strategy reduces the utilities of the hardware resources and also accelerates calculations. This unified calculation methodology is designed to adapt inputs of diverse bit widths, seamlessly decoding them back into their original format after the CRT calculation, thereby significantly enhancing the throughput of calculations. We hope this work advances privacy-preserving computing in resource-constrained environments.

DOI: 10.1145/3676536.3698872

Authors: Tianyou Bao, Pengzhou He, Jiafeng Xie

Institution: Villanova University

Language: English

Topic: Cryptography and Data Security

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Funders: National Science Foundation, National Institute of Standards and Technology

Awards: G4508093486, G4935221308, EFFICIENT HARDWARE IMPLEMENTATION OF LATTICE-BASED POST-QUANTUM CRYPTOGRAPHY, SaTC: CORE: Small: Fast Algorithm Originated Fault Detection Scheme for Ring-LWE based Cryptographic Hardware, G8921104757

OpenAlex ID: W4409282590

FWCI: 0.294

Cites: 25

Cited by: 1

Related to: 10