Research On Modeling Method Of Trust Attack Based On Object-oriented Petri Net

The problem of network security,such as online fraud,network attacks,is increasingly serious.Users who trade on the Internet and share resources may suffer trust attacks at any moment.Therefore,it is necessary to simulate trust attack scenarios and predict the attackers' attack behaviors.Starting from the perspective of trust attack modeling and path evaluation,this paper proposes Trust Attack based on Object-oriented Petri net model(TAP),Trust Attack based on Object-oriented Fuzzy Petri net model(TAFP)and Coordinated Trust Attack Petri net model(CTAP)to solve the problems existing in the research.The main results are as follows:Firstly,on the basis of trust attack graph,TAP is put forward.This model dynamically describes the attack behaviors among components of trust entity object.The trust attack relation between trust entity objects and components is shown in the Petri net,simulating the collaborative work between the components.A trust attack path inference algorithm is proposed to make the attackers knowing the changes of harmfulness when attacking the components,thus they can decide the attack direction.The maximal threat path which can arrive at an attack goal can be searched out in the end.An experiment prove the correctness of the model.Then,in order to solve the problem of fuzziness that exists in the trust attack modeling process,TAFP is proposed.This model introduces the Fuzzy Petri Net and fuzzily processes parameters,which have improved the description ability of model.The fuzzy inference rules and ant colony genetic hybrid path finding algorithm can quickly and efficiently get the fuzzy attack paths that meet the requirement of attack.Then,using fuzzy indicators,such as fuzzy reliability,fuzzy complexity,fuzzy truthdegree,evaluates the path.The usability of the model is confirmed through an example.Finally,CTAP is proposed for coordinated attack.Based on the Trust Attack Petri net model,Time Petri net and Colored Petri net are introduced into this model.Dividing the attack project into stages,coloring the token to distinguish attackers from different areas,using fuzzy time stamp and fuzzy time delay to control the whole trust attack process,the optimal attack path of each project can be found through coordinated trust attack inference algorithm.Then,combining threat degree of projects and real-time degree evaluates the advantage and disadvantage of every project.At last,validation is carried out by means of example.