Delay-tolerant Network Routing Algorithm Based On Social Network Analysis

Internet is a great success. On which the TCP/IP protocol running on the Internethas been the actual standard for network interconnecting. With the development ofscience and technology, new type of networks began to emerge, such as sensornetworks, deep space communication networks, which are characterized with high linkdelays, high link error, frequent disconnection, limited nodes resources, etc. Some of thebasic Internet architecture key assumptions can hardly be fulfilled. Therefore,interconnecting these new networks with TCP/IP encountered great difficulties. Wecalled the internetworks of these new networks Delay-Tolerant Network (DTN).In this paper, firstly we analyze the limitations of TCP/IP in application to DTN,elaborating the difference between DTN and the Internet and explaining the randomrouting algorithm and knowledge-based algorithms. Then we introduce the basics ofsocial network analysis, mainly illustrating the knowledge of centrality, similarity, andcluster analysis and we probe into the possibility of social network analysis inapplication of DTN routing algorithm. Finally, by synergetic analysis of DTN andsocial network, a new DTN routing algorithm based on centrality, similarity andclustering analysis is implemented, which exhibits better delivery ratio and overloadratio. This algorithm is finally achieved by the ONE simulation tool. Simulation resultindicates that this algorithm can afford a variety of excellent performances by screeningand matching different features.