计算机网络病毒传播的概率模型

Probabilistic Model for Virus Spread in Computer Networks

研究网络拓扑对病毒传播的影响。提出在双星图上计算固定概率(病毒成功蔓延整个网络的概率)的数值方法,并成功将其转化为求解线性方程组的问题。通过仿真发现:1)双星网络越不平衡,病毒越容易在其中蔓延;2)让网络规模趋于无穷大,固定概率收敛。固定概率还可以简单地作为判断双星网络是否安全可信的参考指标。

This paper is concerned with the way in which a network topology affects computer virus spread. A numerical method is presented for finding the fixation probabilities, i. e. , the probabilities of the event that virus could successfully spread over the entire network, on a two- star diagraph （TSD） by converting this task to solving a linear system. By numerical simulation, the following phenomena are observed： 1 ） The more unbalanced a TSD is,the more likely virus is to spread over it. 2） Let the network size tend to infinity,the fixation probability on a TSD converges. Moreover, the fixation probability can simply be used to assess the safety and reliability of a two-star network.