Location Privacy Protection Method Based On User Similarity And Incentive Mechanism

With the continuous development of Io T technology,life relies on location information to provide more and more services,which provides convenience in daily life.The reliance on location services leads users to ignore the information leakage problems.The most obvious problem is that users submit detailed location information independently leading to the leakage of location information.In addition,the associated information generated by the user's repeated location information submission can be inferred from the user's sensitive personal information,leading to threats to his personal and property security.In the current technical context,location privacy protection methods use anonymization methods widely to generalize a detailed location information into an anonymous region to reduce the risk of leakage.The following problems exist in the traditional anonymization method: improper selection of dummy locations leads to an increased chance of leakage of real location information;the leakage of large amount of user information due to the attack problem of third-party anonymous servers.This thesis proposes the following solutions to the above problems.1.Aiming at the availability of anonymous location points in false location selection methods,this thesis proposes a dummy location selection method based on the similarity of historical location.Firstly,the false location selection method based on user similarity improves the correlation between the false location information in the location anonymity set and resists the precise information leakage caused by the attacker's background information screening.Also the similarity scheme based on hierarchical graph is used in the similarity problem,which improves the correlation of semantic location information among similar users compared to methods such as cosine similarity,thus improving the degree of obfuscation in the location anonymization set.Finally,a noise injection mechanism based on differential privacy is applied to the above formed location anonymous set to improve the security of the location anonymous set as well as anti-attack to avoid attackers reasoning about the location information in the anonymous set and thus revealing the real location information of users.Through the verification on the data set,the above method improves the anonymity of the user's location information within an acceptable time delay and ensures the service efficiency of the location information after anonymization.2.Aiming at the security problem of information leakage in anonymous servers,this thesis proposes a location privacy protection method combining group intelligence perception system and incentive mechanism.The system architecture based on group wisdom awareness decentralizes the anonymization process,so that the anonymization operation is shared by the users participating in the anonymization process.In the anonymization process,the participating users submit their own information for anonymization assistance.At the same time,in order to increase the motivation of participating users,an incentive mechanism based on reputation value is proposed.With the incentive mechanism,the motivation of the participating users to submit information and the accuracy of the data is increased through rewards,and the risk of the participating users leaking the anonymization set data based on their own interests is reduced through penalties.By verifying on the data set,the result shows that under the premise of effective protection of user detailed information,user participation enthusiasm has been effectively improved,and the anonymous success rate has increased by10%-17% on average.