Skip to content

Revisiting Homomorphic Encryption Schemes for Finite Fields

Authors: Kim, Polyakov, Zucca
Year: 2021
Venue: ASIACRYPT 2021

type: scheme collection: catalogue, schemes status: auto

Links: ePrint 2021/204

Summary

Auto-extracted summary (unreviewed)

Kim, Polyakov and Zucca revisit BGV and BFV, proposing an improved BFV variant whose modified encryption sharply reduces noise growth (making it slightly better than BGV even for large plaintext moduli), a lower-complexity BFV homomorphic multiplication with lazy scaling, an improved RNS decryption, and a more usable BGV variant. Implemented in PALISADE, the best BFV variant is up to 4x faster in homomorphic multiplication than prior RNS BFV for deep computations; BFV wins for small plaintext moduli while BGV is faster for intermediate and large moduli.

Abstract

The Brakerski-Gentry-Vaikuntanathan (BGV) and Brakerski/ Fan-Vercauteren (BFV) schemes are the two main homomorphic encryption (HE) schemes to perform exact computations over finite fields and integers. Although the schemes work with the same plaintext space, there are significant differences in their noise management, algorithms for the core homomorphic multiplication operation, message encoding, and practical usability. The main goal of our work is to revisit both schemes, focusing on closing the gap between the schemes by improving their noise growth, computational complexity of the core algorithms, and usability. The other goal of our work is to provide both theoretical and experimental performance comparison of BGV and BFV. More precisely, we propose an improved variant of BFV where the encryption operation is modified to significantly reduce the noise growth, which makes the BFV noise growth somewhat better than for BGV (in contrast to prior results showing that BGV has smaller noise growth for larger plaintext moduli). We also modify the homomorphic multiplication procedure, which is the main bottleneck in BFV, to reduce its algorithmic complexity. Our work introduces several other novel optimizations, including lazy scaling in BFV homomorphic multiplication and an improved BFV decryption procedure in the Residue Number System (RNS) representation. We also develop a usable variant of BGV as a more efficient alternative to BFV for common practical scenarios. We implement our improved variants of BFV and BGV in PALISADE and evaluate their experimental performance for several benchmark computations. The experimental results suggest that our BGV implementation is faster for intermediate and large plaintext moduli, which are often used in practical scenarios with ciphertext packing, while our BFV implementation is faster for small plaintext moduli. 1 Contents
Source: pdf

Extracted details

Field Value Source Confidence
contribution improved BFV encryption with significantly reduced noise growth llm 0.85
contribution modified BFV homomorphic multiplication with lower algorithmic complexity and lazy scaling llm 0.85
contribution improved BFV decryption in the HPS RNS variant llm 0.85
contribution more usable BGV variant for practical scenarios llm 0.85
contribution theoretical and experimental comparison of BGV and BFV in the same library (PALISADE) llm 0.85
schemes BFV llm 0.85
schemes BGV llm 0.85
libraries PALISADE llm 0.85
tasks exact HE arithmetic over finite fields llm 0.85
tasks binary tree multiplication llm 0.85
tasks polynomial evaluation llm 0.85
tasks inner product llm 0.85
interaction non-interactive FHE llm 0.85
code url http://github.com/ldsec/lattigo (URL found in the full text) regex 0.50
code url https://github.com/Microsoft/SEAL (URL found in the full text) regex 0.50
llm relevance low llm 0.85
key result homomorphic multiplication speed-up: up to 4x (best BFV variant vs prior state-of-the-art RNS BFV, deeper computations) llm 0.85
key result noise growth: improved BFV somewhat better than BGV (contrasting prior results favouring BGV at large plaintext moduli) llm 0.85
key result scheme comparison: BFV faster for small t; BGV faster for intermediate/large t (PALISADE benchmarks) llm 0.85

System card: Revisited BGV/BFV for finite fields

Group 4. Alternative FHE schemes
Family RNS BGV/BFV
Arithmetic Exact integers
Interaction Non-interactive
Model scale -
Status Widely used
Relation to the SoK Modern BGV/BFV variants implemented in OpenFHE.
Note Lower noise and faster RNS BFV/BGV multiplication.
Code https://github.com/openfheorg/openfhe-development
Systems page Revisited BGV/BFV for finite fields

Keyword mentions

Counts of vocabulary terms in the full text (first page of occurrence in brackets).

  • scheme: BFV 265 [p.1], BGV 203 [p.1], CKKS 4 [p.4], FHEW 4 [p.4], RSA 3 [p.30], TFHE 2 [p.4], Leveled / somewhat HE 2 [p.9]
  • library: PALISADE 11 [p.1], HElib 7 [p.8], Microsoft SEAL 5 [p.6], Lattigo 2 [p.6]
  • model: Logistic regression 1 [p.31]
  • technique: Key switching 84 [p.2], Modulus switching 49 [p.3], RNS 47 [p.1], Relinearization 12 [p.5], Bootstrapping 11 [p.4], NTT 11 [p.13], SIMD packing 3 [p.1]
  • assumption: RLWE 8 [p.4], LWE 8 [p.5], Lattice-based 4 [p.29]
  • topic: Homomorphic encryption 27 [p.1], Benchmark 7 [p.1], Open source 2 [p.29]

Files

Path (under papers/) Collection Pages Size SHA-256 Note
references/catalogue/4-051_revisiting-homomorphic-encryption-schemes-for-finite-fields.pdf catalogue 46 535.4 KB 3e77f5330fec
references/schemes/s15-2_eprint-2021-204.pdf schemes 46 535.4 KB 3e77f5330fec byte-identical to references/catalogue/4-051_revisiting-homomorphic-encryption-schemes-for-finite-fields.pdf

Database key: revisiting-homomorphic-encryption-schemes-for-finite-fields · created 2026-11-05 · updated 2026-11-05 · edit with python3 tools/paperdb.py set revisiting-homomorphic-encryption-schemes-for-finite-fields COLUMN VALUE