In order to address the shortcomings of traditional anonymity network anonymity evaluation methods,which only analyze from the perspective of the overall network and ignore the attributes of individual nodes,we propos...In order to address the shortcomings of traditional anonymity network anonymity evaluation methods,which only analyze from the perspective of the overall network and ignore the attributes of individual nodes,we proposes a dynamic anonymity model based on a self-built anonymous system that combines node attributes,network behavior,and program security monitoring.The anonymity of evaluation nodes is assessed based on stable intervals and behavior baselines defined according to their normal operating status.The anonymity of the network is evaluated using an improved normalized information entropy method that refines anonymity evaluation to the anonymity of each node and expands the dimensionality of evaluation features.This paper compares the effectiveness of our proposed method with static framework information entropy and single indicator methods by evaluating the degree of anonymity provided by a self-built Tor anonymous network under multiple operating scenarios including normal and under attack.Our approach utilizes dynamically changing network anonymity based on multiple anonymous attributes and better reflects the degree of anonymity in anonymous systems.展开更多
基金supported by the Tianjin Education Commission Research Program Project No.2019KJ024.
文摘In order to address the shortcomings of traditional anonymity network anonymity evaluation methods,which only analyze from the perspective of the overall network and ignore the attributes of individual nodes,we proposes a dynamic anonymity model based on a self-built anonymous system that combines node attributes,network behavior,and program security monitoring.The anonymity of evaluation nodes is assessed based on stable intervals and behavior baselines defined according to their normal operating status.The anonymity of the network is evaluated using an improved normalized information entropy method that refines anonymity evaluation to the anonymity of each node and expands the dimensionality of evaluation features.This paper compares the effectiveness of our proposed method with static framework information entropy and single indicator methods by evaluating the degree of anonymity provided by a self-built Tor anonymous network under multiple operating scenarios including normal and under attack.Our approach utilizes dynamically changing network anonymity based on multiple anonymous attributes and better reflects the degree of anonymity in anonymous systems.
