Principal Component Analysis over the Boolean Circuit Within TFHE Scheme

Hyun Jung Doh, Joon Soo Yoo, Mi Yeon Hong, Kang Hoon Lee, Tae Min Ahn, Ji Won Yoon

Research output: Chapter in Book/Report/Conference proceedingConference contribution


In today’s information-driven world, the need to protect personal data while maintaining efficient data processing capabilities is crucial. Homomorphic Encryption (HE) has emerged as a potential solution, allowing secure processing of encrypted information without compromising privacy. However, current HE schemes suffer from slow processing speeds, especially when dealing with high-dimensional data. This paper focuses on leveraging the PCA technique within the Fast Fully Homomorphic Encryption over the Torus (TFHE) scheme to optimize the speed of subsequent algorithms. TFHE offers the advantage of enabling the homomorphic implementation of any circuit but suffers from extensive execution time. We present tailored PCA algorithms for TFHE, utilizing the power method and eigen-shift techniques to extract eigenvalues and eigenvectors. These techniques provide efficient solutions for performing PCA computations within the TFHE framework. By designing a dedicated PCA circuit using TFHE’s fundamental homomorphic gates, we achieve efficient evaluation times for PCA. The performance analysis shows execution times of 3.42 h for a 16-bit dataset and 12.22 h for a 32-bit dataset, with potential for further improvement.

Original languageEnglish
Title of host publicationInformation Security Applications - 24th International Conference, WISA 2023, Jeju Island, South Korea, August 23–25, 2023, Revised Selected Papers
EditorsHowon Kim, Jonghee Youn
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages13
ISBN (Print)9789819980239
Publication statusPublished - 2024
Event24th International Conference on Information Security Applications, WISA 2023 - Jeju Island, Korea, Republic of
Duration: 2023 Aug 232023 Aug 25

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14402 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference24th International Conference on Information Security Applications, WISA 2023
Country/TerritoryKorea, Republic of
CityJeju Island

Bibliographical note

Publisher Copyright:
© 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.


  • Homomorphic Encryption
  • Principal Component Analysis
  • TFHE

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Principal Component Analysis over the Boolean Circuit Within TFHE Scheme'. Together they form a unique fingerprint.

Cite this