Anonymous Communication And Leak Detection Mechanism For Privacy Protection

In recent years,frequent disclosure of privacy breaches has caused varying degrees of impact on society and individuals.On the one hand,the TCP / IP protocol requires the datagram header protocol to be transmitted in the network in plain text,and the attacker can still perform traffic analysis attacks based on the unencrypted header information even if the data packet payload of the communication is encrypted in advance.An attacker can identify the originator and the terminator of the session by examining the source address and destination address of the packet header.Then,by analyzing the communication modes of both parties,the attacker can guess the current activity with a high probability.On the other hand,because Android and IOS system for application permissions are based on coarse-grained,that is,the user is responsible for the application's own permissions.However,the average user lacks awareness and professional knowledge of the privacy disclosure of the mobile platform,making the mobile platform a high-risk area for privacy disclosure.In view of the leakage of user privacy and privacy information,many existing schemes propose anonymous communication system model and mobile application detection mechanism.However,for the design of anonymous communication system model,most solutions are difficult to balance the relationship between anonymity and performance.In addition,the existing mobile application privacy leak detection mechanism has high time overhead and can not guarantee the accuracy of the detection result.Taking the existing shortcomings at work into consideration,the specific works of this thesis are as follows:(1)A Twice adjustment routing algorithm(TARA)is designed to construct anonymous communication path,which not only guarantees the anonymity of the communication,but also ensures the best performance of the constructed path.Aad a property of the path participation amount for online active node to record the actual number of times that the current node participates in constructing an anonymous path and uses it as a basis for selecting nodes to effectively balance the load of the online nodes.Introduce a HOOK policy,which replaces the neighbors of the nodes that are about to go offline,effectively reducing the number of reconstruction of anonymous communication paths and avoiding the probability and statistics attacks of malicious nodes.(2)Propose a concept of service binding used for binding a minimum privacy permission set for each service provided by the application.By using the concept of service binding,the static analysis stage only tracks sensitive data flows between APIs that apply for service-bound privacy,which can effectively improve the detection of privacy-sensitive applications.In this thesis,static analysis is combined with the machine learning algorithm effectively to abstract the path characteristics of the static data stream into eigenvectors.The machine learning algorithm is further used to classify the learning paths and improve the accuracy of the test results.(3)Design and implement a privacy protection prototype system that can hide user's privacy behavior and privacy data.The anonymous communication function is based on the P2 P network architecture and uses a rerouting method to send a message,effectively hiding the relationship between the two parties in the communication.The mobile application privacy leak detection function can detect the security of a specific application service of a mobile application and can provide guidance for the application of service-related privacy permissions that users need to use.After experimental verification and security analysis,the proposed scheme of anonymous communication system has better anonymity and performance.The detection mechanism of mobile privacy leak has faster detection speed and higher accuracy than the existing mechanisms.