Paper 2017/430

Improving TFHE: faster packed homomorphic operations and efficient circuit bootstrapping

Ilaria Chillotti, Nicolas Gama, Mariya Georgieva, and Malika Izabachène

Abstract

In this paper, we present several methods to improve the evaluation of homomorphic functions, both for fully and for leveled homomorphic encryption. We propose two packing methods, in order to decrease the expansion factor and optimize the evaluation of look-up tables and random functions in TRGSW-based homomorphic schemes. We also extend the automata logic, introduced in [19, 12], to the efficient leveled evaluation of weighted automata, and present a new homomorphic counter called TBSR, that supports all the elementary operations that occur in a multiplication. These improvements speed-up the evaluation of most arithmetic functions in a packed leveled mode, with a noise overhead that remains additive. We finally present a new circuit bootstrapping that converts TLWE into low-noise TRGSW ciphertexts in just 137ms, which makes the leveled mode of TFHE composable, and which is fast enough to speed-up arithmetic functions, compared to the gate-by-gate bootstrapping given in [12]. Finally, we propose concrete parameter sets and timing comparison for all our constructions.

Metadata
Available format(s)
PDF
Category
Public-key cryptography
Publication info
Preprint. MINOR revision.
Keywords
FHEleveledbootstrappingLWEGSWpackingweighted automataarithmetic
Contact author(s)
ilaria chillotti @ uvsq fr
History
2017-05-22: received
Short URL
https://ia.cr/2017/430
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2017/430,
      author = {Ilaria Chillotti and Nicolas Gama and Mariya Georgieva and Malika Izabachène},
      title = {Improving {TFHE}: faster packed homomorphic operations and efficient circuit bootstrapping},
      howpublished = {Cryptology {ePrint} Archive, Paper 2017/430},
      year = {2017},
      url = {https://eprint.iacr.org/2017/430}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.