Secure network coding

Ning Cai, Raymond W. Yeung

2003 · 452 citations · 4 references

Concepts

TL;DR

Network coding lets nodes encode incoming data, generalizing switching and enabling full network capacity for multicasting. The paper proposes a new model that integrates network coding with information security. The model specifies subsets of links that a wiretapper may access, ensuring that any single subset reveals no message information, and it subsumes secret sharing. The authors construct secure linear network codes whenever a particular graph‑theoretic sufficient condition holds.

Abstract

Recent work on network coding renders a new view on multicasting in a network. In the paradigm of network coding, the nodes in a network are allowed to encode the information received from the input links. The usual function of switching at a node is a special case of network coding. The advantage of network coding is that the full capacity of the network can be utilized. In this paper, we propose a new model which incorporates network coding and information security. Specifically, a collection of subsets of links is given, and a wiretapper is allowed to access any one (but not more than one) of these subsets without being able to obtain any information about the message transmitted. Our model includes secret sharing as a special case. We present a construction of secure linear network codes provided a certain graph-theoretic sufficient condition is satisfied.

References

4