Secure Network Coding over Small Fields

Abstract

In the paradigm of network coding, the information-theoretic security problem is encountered in the presence of a wiretapper, who has capability of accessing an unknown channel-subset in communication networks. In order to combat this eavesdropping attack, secure network coding is introduced to prevent information from being leaked to adversaries. For any construction of secure linear network codes (SLNCs) over a wiretap network, the based field size is a very important index, because it largely determines the computational and space complexities, and further the efficiency of network transmission. Further, it is also very important for the process of secure network coding from theoretical research to practical applications. In the present paper, we proposed new lower bounds on the field size for constructing SLNCs, which shows that the field size can be reduced considerably without giving up any security and capacity. In addition, since the obtained lower bounds depend on network topology, how to efficiently determine them is another very important and attractive problem for explicit constructions of SLNCs. Motivated by a desire to develop methods with low complexity and empirical behavior to solve this problem, we propose efficient approaches and present a series of algorithms for implementation. Subsequently, the complexity of our algorithms is analyzed, which shows that they are polynomial-time.