Weakly Secure Network Coding

In this work we consider the problem of secure data transmission on an acyclic multicast network. A new information theoretic model for security is proposed that defines the system as secure if an eavesdropper is unable to get any “meaningful” information about the source. The “wiretap” network model by Cai and Yeung in which no information of the source is made available to the eavesdropper is a special case in the new model. We consider the case when the number of independent messages available to the eavesdropper is less than the multicast capacity of the network. We show that under the new security requirements communication is possible at the multicast capacity. A linear transformation is provided for networks with a given linear code to make the system secure. The transformation needs to be done only at the source and the operations at the intermediate nodes remain unchanged. We also show that if random coding scheme is used the probability of the system being secure can be made arbitrarily close to one by coding over a large enough field.