PROPAGATION OF COMPUTER VIRUS AND PATCH DISSEMINATION

Abstract

This thesis investigates some mathematical models to analyze computer virus propagation and patch dissemination dynamics. These models incorporate both constant and time-dependent control strategies. Deterministic models are mathematical models that predict the behavior of a system based on a set of fixed parameters and initial conditions. They assume that the systems behavior is predictable and can be accurately described by a set of equations. In the context of computer virus propagation, deterministic models can be used to predict how a virus will spread through a network and how effective different control strategies will be in controlling or eradicating the virus. A node-based model is also developed in this thesis. The model is based om continuous Markov chain equation. Such models explicitly considers individual nodes and the topography of network. The thesis is divided into seven chapters. The network-related dynamical problems for virus propagation and patch distribution were investigated by using mathematical and numerical computations as follows: The first is introduction chapter that provides literature review, basic mathematical concepts and overview of the thesis. The second chapter deals with a SLBPS dynamic model for capturing the propagation of viruses and dissemination of patches. The effect of latency is considered in the infection. Two thresholds, the basic reproduction number R0 for infection and for patching Rp are computed. Further, it is observed that the patching, patch invalidation and auto-curability rate of a latent node significantly affect the virus dynamics even though the value of R0 exceeds unity. The bifurcation diagram between two reproduction numbers R0vs.Rp is drawn to show the stability region of four equilibrium points.