Anonymous node location based on clustering

Abstract

With the rapid development of network technology, the network’s scale is increasing and becoming more complex. It is imperative to understand the structure of the network for network management and network security. Therefore, the construction of the routing topology has become a top priority. Routing topology analysis is already a relatively mature technology, but anonymous nodes’ positioning has not received enough attention. Nowadays, the primary solutions to anonymous nodes are graph analysis and tomographic tree construction. Based on these two methods, this paper proposes a rough location of anonymous nodes based on the method of three-level clustering. They are the K-means clustering of data before anonymous identification, hierarchical clustering of feature values in anonymous identification, and central point calculation based on clustering ideas during anonymous positioning. Experiments have proved that performing the first-level K-means clustering before anonymous node identification can effectively improve anonymous positioning accuracy. And the three-layer clustering idea proposed in this paper can roughly get the location of anonymous nodes based on reducing the complexity of the original method.