Research On Key Technologies Of Anonymity Evaluation In Anonymous Networks

Because of the openness of the Internet,there is a risk of the leakage of personal privacy information of users.In order to protect users' private information,the anonymous network technologies come into being.Anonymity is often used to reflect the privacy protection capabilities of anonymous networks,and achieving high anonymity is the ultimate goal of anonymous networks.Therefore,the quantitative evaluation of anonymity of anonymous networks has become the research focus in academic and technical circles in recent years.From the perspective of network anonymity assessment,detection,confrontation and so on,this paper studies traditional methods such as the measurement methods based on information theory,the evaluation methods based on the set,the evaluation methods based on evidence and matrix theory,etc.In view of the shortcomings of current anonymity evaluation technologies,the paper proposes an anonymity evaluation technology that combines the theories of effective anonymous set and probabilistic graphical models.It comprehensively considers the impact of anonymous set and awarding nodes on anonymity.The research contents mainly include:(1)The paper investigates and analyzes the recent papers and research results on anonymous networks and anonymity evaluation technologies,and summarizes the development of anonymous networks and anonymity evaluation technologies;it is found that the anonymity evaluation technologies based on set theory and matrix theory have the problem in insufficient evaluation of the effect of forwarding nodes on anonymity,and the performance of the anonymity evaluation technologies based on information theory and evidence theory are strongly related to the accuracy of traffic detection.(2)An anonymity evaluation method combining the effective anonymous set and the probabilistic graphical model is proposed.The method uses the probabilistic graphical model to model the anonymous network,establishes the mapping relationship between the message and the anonymous set in the anonymous network with exact inference,and uses the degree of anonymity to represent the anonymity of the network.Firstly,the paper determines the anonymous set by traffic identification.In order to obtain the effective anonymous set,an improved traffic identification method based onmultiple dimensions is used to detect the set.Then,by the entropy weight method,the information of the nodes,including the bandwidth and the running time,is introduced into the calculation of the forwarding probability.Finally,the calculation is substituted into the formula of the information entropy to obtain the anonymity of the anonymous network.(3)The paper designs an anonymous evaluation system based on the above results.Firstly,this paper experiments on the improved traffic identification method based on multiple dimensions.The results show that the accuracy rate of this method to determine the effective anonymous set is 97.6%,which is 5.9% higher than the traditional method.At the same time,the system is deployed in a real network environment.Compared with the traditional anonymity evaluation method,the degree of anonymity is less.When the size of the effective anonymity set is 10,the degree of anonymity of the traditional anonymity evaluation method is 3.3,while the degree of anonymity of the improved method fluctuates around 3.1.At the same time,when the configuration information of the forwarding node changes,its anonymity also changes.Therefore,the anonymity evaluation technology proposed in this paper can better reflect the real network situation when evaluating the network,and it is of great significance for the subsequent research in related fields such as network anonymity and security protection.