Efficient multicast stream authentication for the fully adversarial network model

Christophe Tartary*, Huaxiong Wang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

15 Citations (Scopus)

Abstract

We consider the stream authentication problem when an adversary has the ability to drop, reorder or inject data in the network. We propose a coding approach for multicast stream authentication using the list-decoding property of Reed-Solomon codes. We divide the data to be authenticated into a stream of packets and associate a single trapdoor hash collision for every λη packets where λ and n are predesignated parameters. Our scheme, which is also joinable at the boundary of any n-packet block, can be viewed as an extension of Lysyanskaya, Tamassia and Triandopoulos’s technique in which λ = 1. We show that by choosing λ and n appropriately, our scheme outperforms theirs in time spent for processing data at the sender and receiver. Our approach relies on the dispersion process as SAIDA and eSAIDA. Assuming that we use RSA for signing and SHA-256 for hashing, we give an approximation of the proportion of extra packets per block which could be processed via our technique with respect to the previous scheme. As example when we process λ = 1000 blocks of 2650 64-byte-packets, the gain of our scheme with respect to Lysyanskaya et al.’s is about 30%.

Original languageEnglish
Pages (from-to)175-191
Number of pages17
JournalInternational Journal of Security and Networks
Volume2
Issue number3-4
DOIs
Publication statusPublished - 2007

Bibliographical note

First published in Information Security Applications. WISA 2005. Lecture Notes in Computer Science, vol 3786, pp 108-125. DOI https://doi.org/10.1007/11604938_9

Keywords

  • Reed-Solomon codes
  • signature dispersion
  • stream authentication
  • THF
  • trapdoor hash function

Fingerprint

Dive into the research topics of 'Efficient multicast stream authentication for the fully adversarial network model'. Together they form a unique fingerprint.

Cite this