Modeling the propagation of topology-aware P2P worms considering temporal parameters

Abstract

The propagation of topology-aware peer-to-peer (P2P) worms in the Internet is a serious threat due to their ability to evade traditional detection mechanisms by means of avoiding revealing any abnormal behavior. Recent increase in the popularity and the usage of P2P systems and applications is exponentially increasing the risk of this threat. The existing models of P2P worms’ propagation consider network topology as an important parameter affecting the propagation process. However, a drawback of these models is that they ignore the infection time lag (i.e. the time taken by worm to infect a host). In this paper, we extend the four-factor worm propagation model and propose timed four-factor model to consider both the network topology and the infection time lag. We have also developed an agent-based simulation model to conduct simulative experiments based on the proposed model. The results of our experiments show that the infection time lag has considerable impact on the attack performance of topology-aware P2P worms. The performance of different peer immunization strategies considering the launch time of immunization strategies and other effective parameters have also been investigated in this paper.