Research On Location Privacy Protection Based On K-Anonymity

With the rapid development of wireless communication technology and global positioning technology,location-based services are widely used,and various mobile applications usher in new opportunities.More and more users are beginning to enjoy the benefits of location-based services through these mobile applications.When using this value-added service,mobile users need to actively provide their own location information and request query content.This process will inevitably cause the user's privacy information to be leaked,posing a great threat to the user's personal and property safety.The further development of location-based services is severely restricted by the problem of location privacy leakage,and has received extensive attention from domestic and foreign researchers.Currently,privacy protection methods for location services mainly include location anonymity technology,false location technology,pseudonym technology and encryption technology.This paper analyzes the existing location privacy protection model based on k-anonymity technology.According to the distance entropy between users in the constructed anonymity area and the difference in request content,this paper proposes an anonymous area construction algorithm based on the ball tree--BT-RCA algorithm,and taking into account that users need to log in to the application through their mobile phone number for continuous query,the BT-RCA algorithm is improved,and the trajectory location protection algorithm based on exchange query under identity authentication--EQ-TPA algorithm is proposed.The main research contents of this paper are as follows:1.Analyzed and compared the current four mainstream location privacy protection models,comprehensively considered the advantages and disadvantages of the models,and selected the k-anonymity protection model that takes into account both security and practicability as the research focus,and introduced the current research status and related basics of the model in detail.2.An anonymous region construction algorithm based on the ball tree--BT-RCA algorithm is proposed.Although the k-anonymity model can reduce the probability of a user being identified by a malicious attacker to 1/k,it has the problems of redundancy in the structure of anonymous regions,concentration of users in the region,and the single query content.In order to resist this risk,a BT-RCA algorithm that takes into account the distribution of users and the number of requested content types is proposed.The algorithm uses the ball tree as a storage structure to improve the efficiency of searching for neighbor users,reduce the redundancy of constructing anonymous regions,and comprehensively consider the distance between users in the region and the difference in request content,construct multiple anonymous regions,and use the proposed entropy calculation method,select the anonymous area with the largest entropy to ensure the uniformity of user distribution and the diversity of requested content in the anonymous area.Simulation results show that the algorithm can effectively reduce the area of anonymity while ensuring user privacy requirements,and can resist multiple attack models,and has a high degree of privacy protection for user location information and requested content.3.A trajectory location protection algorithm based on exchange query under identity authentication--EQ-TPA algorithm is proposed.In order to protect the privacy of users from the root cause,the BT-RCA algorithm that only protects user location information and query content is improved,and an EQ-TPA algorithm is proposed.When a user logs in,a trusted third-party server is used to divide the data.The server saves the user's mobile phone number,and the location service provider saves data that has nothing to do with the user's real identity.When a user submits a query,a candidate collaborative user area is constructed according to the user's privacy requirements,and the distance entropy and request content entropy calculation methods are used to select the user with the largest overall entropy as the best collaborative user,and the ID of the user and the best collaborative user The conversion allows the two parties to exchange query information with each other,which not only protects the mobile phone number representing the user's true identity,but also protects the user's location information and requested content information.In the user inquiry process,all messages are encrypted,even if the message is intercepted,the user's private information will not be leaked out.Experimental simulation results show that the algorithm can effectively reduce the system response time,improve the protection of the user's trajectory position,and can resist a variety of attack models,and has a high degree of privacy protection for the user's real identity,location information and requested content information.