De-anonymous and anonymous technologies for network traffic release

Abstract

With the rapid growth of data, the network traffic data is of great significance for the research and analysis. Through research and real events, anonymous network traffic is susceptible to de-anonymity attacks. Therefore, the release of network traffic need to consider the existence of de-anonymization attacks, and balance the privacy and utility of data. On the one hand, we summarize the anonymous technologies of network traffic, list some traffic anonymity methods. on the other hand, we analyze the anonymous strategy of network flow against de-anonymous attacks. Based on the research on de-anonymization attacks, this paper divides the de-anonymization method into three categories from the dimension of inferring attack object: restoring the network topology graph, inferring the host behavior, inferring the node and edge information. Specifically, we analyze the implementation methods of these three types of de-anonymization attacks respectively. In connection with the network traces anonymity method, we analyze the confrontation strategy of the above de-anonymity attacks.