A High-Throughput and Low-Complexity Secure Linear Network Coding Protocol

A new scheme providing security against passive attackers in linear network coding is proposed. Throughput efficiency, low algorithm complexity, and high adaptability/applicability are the major design factors that are considered in this security protocol. A bijective permutation map defined over the code field is utilized to generate the randomness that is required for masking the plain information symbols. The input arguments of the permutation map are some of the plain data symbols and one random symbol that is chosen by the source node. It is shown that as long as the attacker does not have access to all the independent channels, he can not obtain any linear combination of the plain information symbols. Reducing data throughput by only one unit compared to the non-secure network code, and avoiding the use of complex transformations such as cryptographic algorithms or hash functions are the main advantages of the proposed security protocol.