Modeling And Analysis Of Two-Worm Interaction In Wireless Networks

Due to the good compatibility, flexible configuration, and facilitate mobility, wireless networks have been widely applied in military, civil, industrial, commerce and environmental monitoring. While wireless network is promoting the development of national economy, it also suffers more and more attacks from malignant worms and their variants. Wireless networks have some different features from wired networks, such as the openness of transmission signal, terminal mobility and the dynamic characteristics of network topology. Thus, a large number of research results on worms in wired networks cannot be directly used in wireless networks. With high controllability and non-destruction, benign worms are becoming a new emergency response technologies. Besides, the present studies on wireless anti-worm are steel rare. Therefore, modeling new worm-anti-worm models in wireless networks, studying worm propagating dynamics, and developing effective security strategies to resist the propagation of malicious worms have fallen into the scope of research. The main contents and contributions of this thesis are listed below.First, address hiding technology has a great influence on the propagation behavior of worms, but the existing wired or wireless network worm propagation models ignored it. This paper establishes a two-worm interaction model by considering the address hiding of hosts, and several other influence factors in wireless networks. We analyze the dynamical behavior of this model, and get the threshold of controlling the spread of malicious worms and give the effective control strategy. The simulation results demonstrate the correctness of our theoretical results and the validity of strategy.Second, the existing worm models in wireless networks ignored the influences of large number of removable devices. By considering this important factor, this thesis builds an extended SIBVRsRi two-worm interaction model. Through the infection dynamics analysis methods, we analyze the stability of this model and get the threshold of controlling malicious worms. The simulation results intuitively reflect the influences of removable devices on the two-worm propagation, and verify the accuracy of our model and strategy.Third, most of the existing worm models were built on the assumption that network topology is fully-connected. However, there is no one model considering the influences of network heterogeneity on the two-worm propagation. Based on the theory that the topology of a wireless network approximately obeys power-law distribution, this paper presents the first worm-anti-worm model in heterogeneous wireless network. Besides, we analyze and compare the worm propagation behaviors under three different immune strategies, which indicates the validity of benign worm strategy. Moreover, the expected final size formula of malicious worms is deduced, from which we know the total number of nodes affected by worms. Finally, the correctness of theoretical analysis is verified by simulations.This paper studies the interactive dynamic behaviors between benign worms and malicious worms in wireless networks. Through comprehensively considering the characteristics of wireless networks and influence factors, we propose and analyze the dynamic behaviors of different worm-anti-worm interaction models. Our research results provide certain theoretical basis to control the large-scale propagation of worms in wireless networks.