Research Of Email Authenticity Identification Based On Game Theory

With the rapid development of Internet, E-mail becomes an important way for people to transmit information on Internet. In order to solve the civil disputes related to E-mail effectively, E-mail authenticity identification is gradually becoming the research emphasis in the field of E-mail forensics. In recent years, there are many domestic and foreign researches on forensics and anti-forensics against, email forgery and authenticity identification and game theory applications. However, the existing researches are confined to the traditional identification methods, and lack of solutions about E-mail forensic antagonism problem. This shortage make it neither helps solve the contradiction between E-mail forgery and identification, nor improves the efficiency of email authenticity identification. Based on a better understanding of the conflicts and contradictions between forgery and detection, this thesis utilizes a mathematical model to describe the essence of email identification. It can solve the problem of counterfeit behavior prediction and optimal detection strategy selection, and also provides decision-making and reference information for E-mail identification. The main work of this thesis is summarized as follows:Firstly, the common email forgery and identification schemes have been studied in this thesis. The forgery schemes include email address spoofing, email data and time spoofing, and email content spoofing, while the identification schemes include email transmission traces analysis, multiple attributes correlation analysis, multiple emails correlation analysis and so on. The researches on forgery and detection strategies can provide auxiliary information for email authenticity identification.Secondly, a classification method for E-mail forgery and detection strategy based on Operational Complexity is proposed in this part. This method takes the mutual exclusivity, integrity, usability and acceptability as important features, and measures the operational complexity of a strategy by computing the time and the resources it involves among the implementation process. Therefore, a cost-benefit quantification framework is presented. Combining the framework with the strategy classification method can provide decision-making and reference information for email authenticity identification, and lay a foundation for the subsequent game model.Finally, based on the relationship between email forger and detector is opposite, interdependent and non-cooperative, an Email Forgery-Detection Game model is built in this thesis. The FDG model is used to describe the relationship and actions between forger and detector, predicts the forgery actions, and selects optimal detection strategy. Moreover, the FDG model is applied to a practical case, and the experimental results demonstrate that the FDG model is feasible and effective.