Skip to Main content Skip to Navigation
Reports

Fully Homomorphic Encryption and Bootstrapping

Abstract : This report is the result of a work done during the SEME (Semaine d'Étude Mathématiques Entreprise) in Rennes (May 2nd-May 6th 2022). The project presented here concerns the company Ravel Technologies and deals with the field of homomorphic encryption for secure data processing. Homomorphic encryption is an encryption that allows users to do computations on encrypted data without first decrypting them. An encryption algorithm must switch with elementary operations, which are at least addition and multiplication. A straightforward application of a homomorphic encryption for the delegation of calculations concerns cloud computing service where there is a need to perform calculations while preserving the confidentiality of the data, e.g. for the medical and banking sectors. Among possible algorithms, those that have become popular over the last ten years are based on the socalled Learning With Errors (LWE) problem, for performance and security reasons. Since this technique introduces noise into the encryption, which can grow during homomorphic computations to the point that later decryption fails, it is necessary to consider a noise reduction method such as bootstrapping.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03676650
Contributor : Rémi LELUC Connect in order to contact the contributor
Submitted on : Tuesday, May 24, 2022 - 10:50:19 AM
Last modification on : Friday, August 5, 2022 - 2:54:52 PM

File

report_SEME_ravel.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03676650, version 1

Citation

Rémi Leluc, Elie Chedemail, Adéchola Kouande, Quyen Nguyen, Njaka Andriamandratomanana. Fully Homomorphic Encryption and Bootstrapping. [Research Report] IRMAR - Université Rennes 1. 2022, pp.1-12. ⟨hal-03676650⟩

Share

Metrics

Record views

48

Files downloads

30