Analysis Of Internet Macroscopic Topology And Virus Spreading

The spreading of virus is becoming a global problem with the rapid development of Internet. It is necessary to have in-depth analysis of the spreading of virus on Internet so that people are able to cope with the spreading of virus. As Internet is a typical complex network, it can be more objectively reflect the real spreading features of virus on Internet when the spreading behaviors of virus on Internet are analyzed from the perspective of Internet macroscopic topology.Research begins with the detailed analysis of characteristic of degree correlation features, and the computation result indicates that the disassortativity of Internet is weakening. In order to analyze and predict how this trends impact on the virus spreading further, we propose an algorithm that can monotonously change the assortativity coefficient of network by edges rewiring on the condition that degree distribution keeps unchanging.Secondly, the virus spreading features on networks with variable assortativity correlation are studied by means of simulation based on the above edges rewiring algorithm. The result of study indicates that the spreading speed of virus is higher in disassortative network than that in assortative network, but the epidemic threshold is also higher. And the result of the analysis about stable infected proportion shows that the relationship between assortative coefficient of network and the stable infected proportion of virus spreading is affected by effective spreading rate.Finally, we propose a new virus spreading model based on the considering that there are differences among different individuals, and the new model depicts the fact that different individuals have different anti-infectious ability in actual networks. Then we simulate the virus spreading on AS-level topology to verify the correctness and feasibility of the model and analyze the spreading features, the result of analysis shows that initial infected rate and initial infected type of nodes don’t affect stable infected rate, but they have much more influence on the spreading speed. So if people are able to control these factors, more valuable time will be kept out to study methods of inhibiting virus spreading. And it also shows that enhancing nodes’resistance and strengthening protection of important nodes can effectively control virus spreading.