Research On The Dynamic Model And Control Strategy Of Bad Information Propagation In Mobile Opportunistic Networks

Opportunistic network is one type of wireless ad-hoc network,in which wireless network devices connect with each other based on encounter opportunity generated by device mobile behavior,and does not require additional maintenance for the connectivity,stability and reliability of network topology configurations.With the pervasiveness of personal mobile devices and the rapidly development of mobile communication technology,the application area of opportunistic network is becoming more and more wide.For example,opportunistic networks are employed to address communication problem outside of base station coverage area,and communication load balancing among several base stations.However,due to the instability of network topology,node mobility,the limit of energy,the uncertainty of transmission distance generated by communication interference,opportunistic network provides a breeding ground for misinformation to generate and propagate.Therefore,how to develop effective and efficient misinformation-combating strategies to prevent the propagation of misinformation becomes a crucial issue.To explore the propagation dynamics of misinformation in opportunistic networks and construct effective and efficient strategies to prevent such propagation,this thesis has systemically and deeply studied the issues in terms of misinformation's propagation dynamic modeling,propagation safety strategies,and balance problems between strategies effects and costs,based on ordinary differential equation model,partial differential equation model,cellular automata,and Pontryagain maximum theory.The studied works mainly includes the following aspects:(1)To study the propagation dynamics of malware in opportunistic network,this dissertation first develops mobile model,which explicitly describes the mobile behavior of malware.Second,this thesis models the temporal-spatial dynamics of malware propagation under mobile conditions,based on epidemic model and differential equation model.Third,our model obtains the threshold which determines whether the malware continuously propagates or not,through making stability analysis for the solution of differential equation.Based on the threshold,then,the malware-countermeasures are developed which can realize the tradeoff between rumor-restraining effect and restraining cost,so that the minimum scale of infected nodes can be obtained with minimum restraining cost,at the end of a certain time.(2)With the increasing threat of rumor propagation on social stabilization and daily life,this thesis studies the propagation dynamics of rumors in opportunistically social network and how to develop effective and efficient countermeasures to restrain rumor propagation.To address these two issues,this thesis mainly focuses on two aspects:first,how to evaluate the influence of network heterogeneity on rumor propagation;second,how to develop optimal countermeasures,considering the tradeoff between cost and effect for proposed countermeasures.For the above problem,this thesis proposes heterogeneous-network based epidemic model which collectively describes the effect of network heterogeneity and different countermeasures on rumor propagation.Through analyzing the existence and stability of equilibrium solution of differential equation system,our model obtains the threshold determines whether rumors continuously propagate or not.Moreover,our model considers two countermeasures and their cost;that is,block rumor at influential spreader and clarify rumor by spreading truth.Last,based on Pontryagain maximum theory,optimal countermeasures are obtained that guarantee that rumors intend to die out with minimum cost,at any time to terminate countermeasures.(3)Considering continuous immunization on opportunistic network virus generally results to high cost and the realistic feasibility of pulse immunization on opportunistic network,this thesis collectively models the mobile features,speed,communication radius,pulse immunization rate,and develops virus propagation-immunization model in opportunistic network based on differential equation model.This thesis analyzes the propagation dynamics of virus in opportunistic network under pulse immunization,and the threshold that determines whether virus will continuously propagate or die out,through exploring the existence and stability of periodic solutions for the proposed differential equation model.Finally,this thesis explores the influence of mobile speed,communication radius,pulse period,and pulse immunization rate on such threshold.Theoretical analysis and simulations all indicate that the strategies proposed herein possess certain theoretical and practical reference values for preventing and controlling the virus propagation in mobile opportunistic network.(4)In order to control the propagation of misinformation and hostile data in adaptive network,this thesis explores the dependable relations between reconnecting-edge strategies.Most of previous researches involving reconnecting-edge strategies are mainly about the random reconnect strategy.Considering the two major factors which determine information propagation behavior in opportunistic network:node degree and node distance,,this thesis proposes a reconnecting-edge strategy based on the shortest path and node degree.Virus propagation model in adaptive complex network was first established according to the cellular automata theory.On this foundation,this thesis also proposes four types of reconnecting-edge strategies:Degree,SP,SP-Degree,and Degree-SP,and systematically explores the influence of reconnecting-edge strategy on malware propagation in Small-World network and Scale-Free network,respectively.A major conclusion is obtained:the SP-degree strategy is more powerful to break the connectivity and transmission in Small-World network;the Degree-SP strategy is more powerful to break the connectivity and transmission in Scale-Free network.This project is funded by the National Natural Science Foundation of China named"Research on theories and methods of temporal-space dynamics about propagation of malicious code in mobile wireless sensor networks"(No.61173094).In the thesis,systematic studies were conducted on the misinformation propagation dynamics in opportunistic networks and effective and efficient countermeasures to restrain misinformation propagation.We proposed different propagation models under different conditions and several types of countermeasures to restrain the misinformation propagation,which can provide theoretical and technical support for the well-designed information propagation model and countermeasure study of opportunistic networks.The results also possess certain practical significance and prospects in mobile wireless environments.