Markov Model of Malicious Code Propagation
Peifeng Wang, Meng Shang, Hui Zhang, JiChao Wang · 2010
In this paper, propagation process of malicious code in computer networks are analyzed by discrete-state Markov model. Computer system without defense mechanism of virus in the networks can be classified: susceptible(S), quarantine(Q), infection(I) and health(H). But the state of computer system is varying. These varieties are only relative to the state at present, and are disrelated to the past state. That is, it is Markovian. So it is appropriate to analysis the propagation process of malicious code in computer networks with Markov model. The model introduced in this paper is a kind of discrete finite-Markovian process. In this method, the transition probability between various state and the one step transition probability matrix can be obtained under given initial, thereby stationary state can be calculated after several step transition. Then the property of all kinds of states is analyzed later. From these the distribution of stationary state has nothing to do with the distribution of inceptive state, and once achieving stationary state, the state don't change any more almost. In this paper the passage time of the stationary state has been gotten based on the property of all kinds of states also, it is very helpful to defense the malicious code.