Showing papers on "Ciphertext published in 1984"
TL;DR: The cryptographic equations relating plaintext, ciphertext, and key-matrix elements in the Hill system are nonlinear equations if the cipher alphabet is unknown and are analyzed by a method of successive eliminations of unknowns from a series of related row-reduced echelon forms.
Abstract: The cryptographic equations relating plaintext, ciphertext, and key-matrix elements in the Hill system are nonlinear equations if the cipher alphabet is unknown. In the case where plaintext is known it is possible to reduce these equations to linear equations by the introduction of a larger set of unknowns. These latter equations are analyzed by a method of successive eliminations of unknowns from a series of related row-reduced echelon forms.
52 citations
01 Jan 1984
TL;DR: This paper proposes a solution to the security problems of field based protection, which allows projections to be performed and individual data elements decrypted or authenticated at the field level.
Abstract: Database encryption and authentication at the field level is attractive because it allows projections to be performed and individual data elements decrypted or authenticated. But field based protection is not usually recommended for security reasons: using encryption to hide individual data elements is vulnerable to ciphertext searching; using cryptographic checksums to authenticate individual data elements is vulnerable to plaintext or ciphertext substitution. Solutions to the security problems of field based protection are proposed.
39 citations
16 Jul 1984
TL;DR: For any fixed R Algorithm S runs to completion in time polynomial in n on all knapsacks with rate Ro>-R, and it successfully breaks at least the fraction \(1 - \frac{{c_R }}{n}\) of such knapsack cryptosystems as n → ∞, where cR is a constant depending on R.
Abstract: This paper gives a performance analysis of one variant of Shamir's attack on the basic Merkle-Hellman knapsack cryptosystem, which we call Algorithm S. Let \(R = \frac{{\# plain text bits}}{{maximum \# cipher text bits}}\) denote the rate at which a knapsack cryptosystem transmits information, and let n denote the number of items in a knapsack, i.e. the block size of plaintext. We show that for any fixed R Algorithm S runs to completion in time polynomial in n on all knapsacks with rate Ro>-R. We show that it successfully breaks at least the fraction \(1 - \frac{{c_R }}{n}\) of such knapsack cryptosystems as n → ∞, where cR is a constant depending on R.
19 citations
01 Jan 1984
TL;DR: The issue of bit security is of even greater importance for the RSA [5] cryptosystem: a positive answer would yield a simple cryptographically secure pseudo random number generator [2, 6, 7].
Abstract: Proving that a public key cryptosystem is bit secure (i.e. proving that guessing even one bit of the cleartext from the cyphertext is hard) is important because it guarantees that the cryptosystem does not leak any partial information. In addition, it also proves that the cryptosystem is suitable for sending 1-bit messages (e.g. a yes-no message). The issue of bit security is of even greater importance for the RSA [5] cryptosystem: a positive answer would yield a simple cryptographically secure pseudo random number generator [2, 6, 7]. With the advent of the RSA chip, it may be a practical generator.
16 citations