# FHE Meeting #3: Paillier encryption

**URL:** <https://forum.anoma.net/t/fhe-meeting-3-paillier-encryption/236>\
**Category:** FHE kindergarten\
**Created:** [September 12, 2023, 1:58pm UTC](https://forum.anoma.net/t/fhe-meeting-3-paillier-encryption/236 "2023-09-12T13:58:40Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![vveiln](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.anoma.net/vveiln/32/60_2.png) [@vveiln](https://forum.anoma.net/u/vveiln)\
**Post date:** [October 23, 2023, 2:25pm UTC](https://forum.anoma.net/t/fhe-meeting-3-paillier-encryption/236/3 "2023-10-23T14:25:52Z")

</div>

A short summary:

- Paillier encryption is based on some fancy [residuosity math assumption](https://en.wikipedia.org/wiki/Decisional_composite_residuosity_assumption) and is additively homomorphic: E(m\_1) \* E(m\_2) = E(m\_1 + m\_2). Provides more or less the same homomorphic functionality as [ElGamal encryption](https://forum.anoma.net/t/fhe-meeting-2-elgamal-encryption/205/3) but is based on a different assumption which might be handy for diversifying security
- CryptDB allows performing computations on encrypted data and uses Paillier encryption as one of the encryption types to support homomorphic queries. CryptDB’s _onion encryption_ that uses multiple types of encryption providing different guarantees allows to find a compromise between:
  - privacy (more operations on encrypted data → less privacy),
  - customizability (does the query require the operation, e.g., ordering of the entries or addition, to be performed at all?),
  - and performance (is it fast enough to be used in real world systems?)

---

_[View the full topic](https://forum.anoma.net/t/fhe-meeting-3-paillier-encryption/236)._