The Epidemics Propagation And Two-phase Dynamic Immunization Strategy On Computer Network

Network viruses are a great threat to the computer networks. The tremendous flows aroused by them will congest the network, and cooperated with hackers they will become more dangerous. Traditional anti virus techniques focus on protection of individual rather than the network, also they nearly could do nothing to stop the denial of service attack aroused by the epidemics. Apparently the network viruses are far beyond us in this race game, we need more effective countermeasures to strike back the viruses.Some epidemics models are proposed to describe the viral propagation; however some of the assumptions in them are not reasonable in the computer network. Computer network viruses can only spread between those nodes which have actual contact, a node will be infected only when some of its neighbors are infective and an infected one will never get recovered without manual immunization.We propose the Dynamic Susceptible-Infectious-Recovered (DSIR) model to overcome such shortcomings; also the immunization delay and the effect of the network topology on the epidemics are introduced. The virus propagation is divided into two phase: in the first phase no immunization exists and the susceptible nodes are infected, in the second phase the immunization is employed, so that the susceptible and infectious nodes are recovered.The static immunization strategies are evaluated based on the DSIR model. Both the analysis and the simulation shows that the efficiency of the targeted immunization decreases sharply with time gap between the vaccines and epidemics considered. Targeted immunization could not suppress the epidemic as efficient as the results got before. We propose a two-phase immunization strategy (TPS) to suppress the computer network epidemics by the spreading of vaccines. During the two phases, the vaccines will go up the degree sequence in phase one and down the sequence in phase two, so the important nodes are protected and the revisit rate of the vaccines is reduced. It is completely decentralized and exploits the power-law distribution in the node degree, and uses local information such as the identities and connectivity of a node's neighbours, but not the target's global position. We also evaluate the efficiency and the cost of the two-phase strategy on the scale free network and hierarchy network.The examination circumstances are created in Windows, and the efficiency test shows: Two-phase strategy can immunized above 90 percent of the nodes in a short unit time, and it can protect the important nodes as a speed close to the fastest Anti-worms; when no immunization delay exists, two-phase strategy can suppress the epidemic as efficiently as the targeted immunization. The immunization cost test shows it will bring less vaccine spreading flows and lower vaccine revisit rate than Anti-worms.