Malware Propagation And Inhibition In Vehicular Ad Hoc Networks

As a high-speed mobile ad hoc network application,vehicular ad hoc network(VANET)has great potential in traffic safety applications such as early warning of traffic accidents and road condition monitoring.The existing researches on malware inhibition methods are mostly based on online social networks,however,there are few researches on the propagation and inhibition of malware in vehicular ad hoc networks.Moreover,most of the research on malware propagation in the vehicle network are based on simply road scenes and randomly vehicle trajectory,such research is lack of scalability and reality.Based on the above characteristics,this thesis mainly carries on the following research: build a vehicle network simulation scene that close to realistic,analyze the topology of vehicle network and the characteristics of the spread of malware in the vehicle network,and the influencing factors of the malware propagation are quantitatively analyzed,moreover,propose a new malware propagation inhibition strategy,compared with the previous suppression strategy,make the research on the propagation in the vehi cle network and its inhibition more realistic and extensible.In order to reflect the authenticity of the simulation experiment,the scene of the 24-hour traffic in Luxembourg city and the moving track of the real vehicle are used as the simulation scenari o.Information propagation based on the SIR epidemic model is realized in the constructed vehicle network model.The nodes in the vehicle network are divided into susceptible nodes,infected nodes,immune nodes.In this thesis,the orthogonal experiment is designed to analyze the factors that affect the spread of malware in the vehicle network,and the concept of information entropy is introduced to select the factor level in orthogonal experiment.The results show that the effective transmission radius and the receiving probability of the vehicle greater effect on the spread of malicious software.Extracting vehicle-specific information at a particular time and constructing an on-board network topology with social attributes,such as node trajectory,node infection status,edges between nodes.Based on the network topology diagram,the betweenness centrality,degree centrality and closeness centrality of the nodes in the network are calculated,and use CRITIC method to calculate the objective weighting coeff icient of each index to get the calculation formula of the node's influence.Finally,maximum influence node target immunity based on Greedy algorithm is proposed,compared with stochastic strategy and target immunity strategy based on the maximum degree nodes.It's concluded that the target immune method based on influence node has the best effect.