Research On The Propagation Model Of Multi-area Worm

With the rapid development of the information technology, regardless of the individualor enterprise, from civilian to military, Internet plays an important role as a bridge ofinformation. Therefore, lots of data to transfer in the network is unavoidable. However, thenetwork is open, information theft is a common accident. If the lawless use of these data, itwill cause damage to individuals and enterprises, even affects the national defense securityand the people’s livelihood. Visible information security is an important concern and preventproblems. While the worm is one of the means used by hackers, can according to the ordinaryuser attack, even for the penetration of the government’s strategy. Therefore, in recent years,scholars patterns make a series of research on their communication, so propagation modelingis a hot issue.At present, research on worm propagation model, include the continuous time modelwith epidemic model as the foundation, and discrete model approximation of discrete timestep, in addition to specific situations of various propagation model. But no much research inmany aspects of the multi-area worms, or their research is not professional enough. Aiming atthe characteristics of worms in the cross region dissemination, based on discrete timeproposes a multi regional worm propagation model. The model of the cross regiondissemination of such occasions have predicts. In this paper, the main research content is:(1) Based on the distribution of vulnerable hosts, area scan focusing factor, scan delayand scanning mechanisms to establish a model of multi-area scanning worms. On this basis,we propose a multi-area worm propagation model (Areas-AAWP model). Under this modelworm propagation of regional selectivity, the effect of infection with regional differences.Analysis of the reasons,it is mainly by the size of the scan rate of the decision, and the fourfactors of the scan rate that affect the underlying causes.(2) On the spread of worm, the correlations between regional scanning behavior isanother important influence factor beyond the size of scan rates. In order to study the correlation between the specific impact on the spread, using the similarity matrix theory, jointhe correlation factor in the Areas-AAWP model, analysis in detail the situation of the modelin different relationship. Theory and experiments show that, under the ideal condition, thegrade of the correlation of a region can decide which worm propagation rates, which led to thedifference degree of infection in multi regions.(3) In order to influence, analysis of network factor in addition to the worm scanning,firstly, using the knowledge of graph theory, construct connectivity relations in multi areainter domain. Secondly, joining the connectivity factor in Areas-AAWP model, analysis theconnectivity of its spread effect in two kinds of scanning case. Finally, the theoretical proofand experimental verification, under two kinds of circumstances, the connectivity relations inmulti area will exert a great influence on the worm propagation rate in a single area andconnected domain.