On privacy homomorphisms

  • Authors:
  • Ernest F. Brickell;Yacov Yacobi

  • Affiliations:
  • Bell Communications Research, Morristown, New Jersey;Bell Communications Research, Morristown, New Jersey

  • Venue:
  • EUROCRYPT'87 Proceedings of the 6th annual international conference on Theory and application of cryptographic techniques
  • Year:
  • 1987

Quantified Score

Hi-index 0.00

Visualization

Abstract

An additive privacy homomorphism is an encryption function in which the decryption of a sum (or possibly some other operation) of ciphers is the sum of the corresponding messages. Rivest, Adleman, and Dertouzos have proposed four different additive privacy homomorphisms. In this paper, we show that two of them are insecure under a ciphertext only attack and the other two can be broken by a known plaintext attack. We also introduce the notion of an R-additive privacy homomorphism, which is essentially an additive privacy homomorphism in which only at most R messages need to be added together. We give an example of an R-additive privacy homomorphism that appears to be secure against a ciphertext only attack.