|
In recent years, the rise of complex network makes people began widespread concern the complexity of network structure and network relationships between behavior. Studying complex networks, research real world complex system topologies and some important statistical properties, master them, these are the key research topics. People on the exploration of complex networks has undergone major changes, complex network theory is also not limited to one subject, one area, but across many different disciplines. Currently, the complex network of research results have been applied to all aspects of life, and promote social development, but there are many problems to be solved. This paper studies the complex network of virus propagation. Since the inception of human society, as head of human infectious diseases has been the enemy. With the development of computer technology, computer viruses have been plaguing people's normal life, cause trouble, it also brings huge losses. Therefore, the spread of the virus is an international research has been the focus of scientists. Transmission of the virus on the study of complex networks can recognize virus propagation characteristics, which can propose appropriate immunization strategy to suppress the virus and reduce losses. Previous models mainly on the right network model to study real-life networks are mostly right network, the study weighted spread of the virus on the network behavior makes more sense. In this paper, three typical weighted network behavior and the spread of the virus immunization strategy research, the main research results are as follows. First, we apply the new mechanism of infection, so that the probability of virus infection and the network connection weights are related, using SIs model virus WANG network model, BBV network model and the network model three kinds ZHU weighted network model of transmission dynamics behavior research. Studies show that: the network size, the average degree of the node, and the node, the average weight of the same conditions, a uniform network virus outbreak on the network than the non-uniform to be fast, steady infection rate, shorter time to reach steady state, the network weight distribution is more uniform, steady infection rate is lower, while the spread of the virus transient process is longer. The initial rate of infected nodes in the network only transient impact viral transmission, the node will not affect the steady-state value infection. Restore the health of infected nodes smaller the probability, the network reaches steady state, the proportion of infected nodes increases. Secondly, we apply and the network connection weights are associated probability of infection, the virus used in the SIR model WANG network model, BBV network model and the network model three kinds ZHU weighted network model to study the propagation dynamics. Studies have shown that the probability of infected nodes removed the greater spread of the virus transient process is shorter, faster time to reach steady state. While using SIR model, to reach the steady state ratio of the nodes on the network disease is very low. WANG ZHU steady state model and a larger model infection, infection process is relatively similar, BBV model steady lowest rate. For not a very serious disease, the lower the probability remove the virus in the network propagation transient time increased significantly. Finally, we use random immunization and objectives based on the maximum weight immunization strategy to study WANG network model, BBV network model and the network model ZHU weighted network model on three kinds of immune effect. Research shows that due to the weight distribution BBV network heterogeneity in BBV network node based on the maximum weight is much better than a random target immune immune to select five percent of generosity node immunization can eliminate the virus. The weight distribution is relatively uniform model and WANG ZHU model, the random node-based immunization method was better than the target weight immunization.
|