Title: Stream cipher based key derivation function

Authors: Chuah Chai Wen; Edward Dawson; Leonie Simpson

Addresses: University Tun Hussein on Malaysia, Queensland University of Technology ' University Tun Hussein on Malaysia, Queensland University of Technology ' University Tun Hussein on Malaysia, Queensland University of Technology

Abstract: A key derivation function (KDF) is used to generate one or more cryptographic keys from a private string together with some public information. To date, most of the existing KDF proposals are designed using hash functions and block ciphers. This paper proposes alternative designs for constructing KDFs using stream ciphers. We show that these designs are secure in relation to the adaptive chosen context information model with single salt (CCS) security model from Krawczyk based on the assumption that the underlying stream ciphers are secure. We simulate instances of these stream cipher based KDFs using three eStream finalists: Trivium, Sosemanuk and Rabbit. The simulation results indicate that the stream cipher based KDFs offer significant efficiency advantages (in terms of running time) over hash function and block cipher based KDFs.

Keywords: KDF; key derivation function; cryptographic key; stream cipher.

DOI: 10.1504/IJSN.2017.083813

International Journal of Security and Networks, 2017 Vol.12 No.2, pp.70 - 82

Received: 04 Apr 2015
Accepted: 06 Apr 2016
Published online: 24 Apr 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article