The Comprehensive Influence of Network Structure, Heterogeneity and Dynamic Behavior on Network Virus Propagation

Abstract

There are two shortcomings in current research on network virus propagation. One is that all virus propagation models assume that the virus can only spread by infecting neighbor nodes. The other is that infection rate is set by a single factor, ignoring important structural information and behavior information of the network. To address these problems, we take the static structure characteristics of the network, dynamic behavior, and heterogeneity into account, and analyze the impact of the three elements on the infection rate. In the actual process of virus propagation, it is not always preferred to infect neighbor nodes. Through our research, we found that the network based on the three elements hides a propagation mechanism with wormhole effect. There is a hidden effective distance between any two nodes. When the susceptible nodes have a shorter effective distance from the source of infection, the infection rate will increase. The node with a shorter effective distance from the source of infection will be preferentially infected. The 'jumping' virus propagation with wormhole effect is presented on the whole network. Based on the classic SIR model, we propose the SIR-SHB model.