Research And Application Of Control Strategy Against Computer Virus Propagation

Computer virus attack is a typical network security event,and its essence is a piece of malicious code.It can not only destroy computer software and hardware facilities,but also encrypt files,tamper with data,and steal secrets,etc.,thus bringing huge economic losses to human society.In recent years,with the popularity of the Internet and the Internet of Things,the destructive power of computer virus has increased,and some viruses can even threaten human lives.In order to effectively curb the spread of computer virus and reduce economic losses,humans need to master the law of computer virus propagation and control strategy.Therefore,the research on control strategy against computer virus propagation will be of practical significance and commercial value.The spreading dynamics of computer virus is an emerging interdisciplinary discipline,whose purpose is to establish dynamic models to fully understand the propagation laws of computer virus and provide theoretical support for proposing cost-effective containment strategies against computer virus.Current research often focuses on controlling the rate of virus-killing by antivirus programs,neglecting the development and transmission of anti-virus programs,which play a key role in controlling the spread of computer virus.In addition,the current research lacks a combination with real industries.In order to make up for this deficiency,this paper is devoted to exploring the research and application of control strategy against computer virus propagation.The main research contents are as follows.(1)Optimal manpower allocation for anti-virus program developmentAnti-virus programs(commonly known as anti-virus software)have a good practical effect in combating computer virus,however,the problem of optimal manpower allocation for developing anti-virus programs has not been effectively solved.This work aims to address the problem mentioned above.First,the optimal manpower allocation problem for the development of single anti-virus program is studied,and a node-level SI(susceptible-infected)single-virus propagation model is established to separately quantify the expected economic loss and risk loss caused by the virus;then we obtain the optimal number of project teams and explore the impact of parameter changes on the optimal number of project teams.Second,we set up a node-level SI dual-virus propagation model,which quantifies the expected economic losses caused by virus 1 and virus 2,respectively.Based on a large number of numerical simulation experiments,the DOWNHILL-DEVELOPMENT algorithm is proposed,and compared with three heuristic algorithms and the random algorithm.Finally,the influence of some parameters on DOWNHILL strategy and its expected total losses are researched.(2)Optimal bandwidth allocation for anti-virus program transmissionOnce the development of an anti-virus program is completed,cyber-security companies need to cost-effectively deliver them to regional servers.This work studies the optimal bandwidth allocation strategy to transmit anti-virus programs.First,we establish a node-level SIPS(susceptible-infected-patched-susceptible)propagation model,and propose a hybrid anti-virus program propagation strategy based on bandwidth(including injection and forwarding),which quantifies the expected economic losses and the expected risk losses,respectively.Second,based on a large number of experiments,the DOWNHILL-TRANSMISSION algorithm is proposed,and compared with the random algorithm.Besides,we also propose the dynamic DOWNHILL strategy.Finally,the influence of some parameters on static DOWNHILL strategy and its expected total losses are researched.(3)Optimal control of virus-killing rateOn the premise that the control cost function is a kind of functions,this paper studies the optimal control problem of virus-killing rate.First,we establish a node-level SLBS(susceptible-latent-bursting-susceptible)propagation model,which quantifies the expected economic losses and the expected total control cost,respectively.Second,the existence of an optimal control is proved and the optimality system is derived.Based on this,we solve three specific examples to show the effect of potential optimal control repair strategies on different networks.Finally,the performance evaluation of the potential optimal control repair strategy is given.(4)Control strategy against computer virus propagation for smart gridIn recent years,smart grid has become the focus of hacking attacks;thus,it attracts widespread attention in academia.The core function of an anti-virus program is to locate and clear computer virus,i.e.,killing virus.The killing rate as a controllable variable plays a key role in curbing the spread of computer virus.This work studies the impact of computer virus attack on operations of smart grid and the potential optimal repair strategy.First,we establish a node-level SI(susceptible-infected)propagation model,evaluating the harm to the grid(the virus will take advantage of false data injection to tamper with meter readings)and the impact of virus propagation rate on the harm.Second,we establish a node-level SIR(susceptible-infected-recovered)propagation model and prove the existence of an optimal control as well as derive the optimality system.On this basis,we solve a specific example on a standard power system platform to demonstrate the effectiveness of the potential optimal control repair strategy.Finally,by comparing with the random algorithm,we further demonstrate the superiority of the potential optimal control repair strategy.