Collaborative Security Diagnosis And Control In Networked Information System

With the rapid development of Internet technology,the progress and development of information technology has become an indispensable strategic resource for the development of our society.Security is one of the most basic characteristics of information-driven systems.It includes interference,anonymity,and Concealment,etc.For information-driven systems,any important information should not be destroyed or obtained by unauthorized users(called intruders).The concept of interference is to avoid any information leakage and prevent intruders from obtaining arbitrary secret information.Most of the existing methods to study the interference of the system by establishing a network model are global diagnostic methods.They have poor flexibility and fault tolerance,and are complicated in calculation.Aiming at the problems existing in the prior works,this thesis provides a method for collaborative security diagnosis of fully distributed subsystems based on Petri nets to analyze and verify strong non-deterministic non-interference(SNNI)and dual simulated strong non-deterministic non-interference(BSNNI).Based on this method,the security of related systems can be diagnosed in real time and efficiently.At the same time,when the system is not secure,you can effectively block information leakage by adding a controller to make the previously unsafe system nodes become A safe.This thesis first proposes a distributed method to analyze the SNNI and BSNNI existing in bounded Petri Nets(PNs).Specifically,a complex large-scale upper-level system is decomposed into multiple small-scale lower-level systems with a simple structure,and the equivalence of behaviors and states between the large-scale upper-level system and the decomposed small-scale lower-level systems is verified.In addition,an important criterion is proposed to explain the relationship between the upper and lower systems of SNNI and BSNNI.Then,two sufficient and necessary conditions are proposed to analyze the SNNI and BSNNI of each subsystem.In the end,the SNNI and BSNNI of a large-scale upper system can be determined by its subsystem through an effective method.That is,when analyzing SNNI and BSNNI,you can pass local analysis without knowing global information.The method proposed in this thesis not only reduces the computational complexity,but also reduces the time complexity.In addition,it verifies the scalability,adaptability,flexibility,and fault tolerance of SNNI and BSNNI analysis in network systems.Finally,the representative examples are compared with existing methods to fully demonstrate the good performance of the distributed method.Secondly,this thesis describes that when a system is verified to be non-SNNI or non-BSNNI through a distributed diagnostic method,a controller is added to make it eventually become SNNI or BSNNI.First,the regional theory method is briefly described.Then,by combining with the definition of SNNI or BSNNI,the general regional theory method is improved to make it suitable for the addition of the controller in this thesis.Then,the subnet for information leakage is calculated based on the improved algorithm.The number of controllers that need to be added and the emission status of the changes,and the SNNI or BSNNI verification of the subnet after the controller is added.Finally,this method is used for distributed control and global control.Compare this The similarities and differences between the two applications,and the advantages and disadvantages of adding controllers to distributed and global systems,and draw a final conclusion.