Modeling the spread of virus in packets on scale free network

Abstract

In this paper, we propose a new model for computer virus attacks and recovery at the level of information packets. The model we propose is based on one hand on the susceptible-infected (SI) and susceptible-infected-recovered (SIR) stochastic epidemic models for computer virus propagation and on the other hand on the time-discrete Markov chain of the minimal traffic routing protocol. We have applied this model to the scale free Barabasi–Albert network to determine how the dynamics of virus propagation is affected by the traffic flow in both the free-flow and the congested phases. The numerical results show essentially that the proportion of infected and recovered packets increases when the rate of infection [Formula: see text] and the recovery rate [Formula: see text] increase in the free-flow phase while in the congested phase, the number of infected (recovered) packets presents a maximum (minimum) at certain critical value of [Formula: see text] characterizing a certain competition between the infection and the recovery rates.