Research On The Privacy-preserving Methods For Location Privacy And Query Privacy

With the rapid development of Internet technology and mobile communication technology,mobile devices can achieve more and more functions,and real-time positioning is one of the more widely used functions.On this basis,various location-based services have been developed rapidly and have been applied in people's livelihood,military affairs,emergency rescue and so on.It has brought great convenience to all aspects of people,so LBS is also considered one of the most promising industries.When users use LBS to query services,they need to submit relevant personal information to LBS service providers,and there are serious security risks to their personal privacy.Because LBS service providers may infer the users' other personal privacy through the users' query request because of curiosity,some malicious attackers will obtain part of the users' privacy by attacking the LBS server and more privacy information by data mining.These security risks increase the users' concern when using LBS services and hinder the healthy development of the LBS market.Therefore,studying the users' privacy protection strategy in the LBS can ensure that users can obtain high quality services under the premise of protecting users' privacy.This is of great significance for users to obtain good service experience and the long-term healthy development of LBS market.According to the research on the existing users' privacy protection methods in locationbased services,it is found that different privacy protection schemes have different degrees of emphasis on the quality of service,the intensity of users' privacy protection and resource overhead in the LBS query process and have not achieved good trade-off among the three,and basically protects the user's location privacy,and rarely considers the user's query privacy.Secondly,in the case of protecting users' location privacy and query privacy at the same time,insufficient consideration is given to the background knowledge that the attacker may have,resulting in the security of the protection scheme not reaching the expected effect.In view of the above problems,this paper makes a thorough study and gives the corresponding privacy protection scheme.The main research work is as follows:(1)This paper first analyzes and compares various existing LBS privacy protection schemes,finds and summarizes the shortcomings of the existing methods,and provides specific and effective solutions through in-depth investigation of different schemes of the same type of methods.(2)A privacy protection k nearest neighbor query method based on service similarity is proposed.Based on the service similarity characteristics of location query and the homomorphism characteristics of Paillier password system,this method realizes the privacy protection of users' location and query content and the accurate query of interest points without relying on trusted third part.At the same time,the disturbance location is generated by constructing a service similarity map,which solves the problem of high query processing overhead of existing methods and ensures the accuracy of query results.The real data set is used to verify the proposed method from the aspects of accuracy and system overhead.The Experimental results and safety analysis show that: compared with the existing methods,the communication complexity of the proposed method is reduced by 33.8% on average,and the query accuracy is improved by 10% compared with the k anonymous scheme,and the privacy protection is higher.Compared with encryption,the time cost is 55 times lower on the basis of both location privacy and query privacy protection.The proposed method effectively guarantees the service quality and reduces the system processing overhead on the basis of both location privacy and query privacy protection.(3)This paper proposes a privacy protection k-nearest neighbor query method against background knowledge inference attacks.This method mainly considers the query probability between different regions and constructs the disturbance position candidate set by greedy algorithm,so that the query probability between the positions of the candidate set is similar,and avoids the attacker combining background knowledge to improve the success rate of attack.And considering the query probability of different interest point types in a certain position,the location close to the number of various types of queries is chosen as the disturbance position,so as to avoid the attacker using background information to screen out some query information.The real data set is used to verify the proposed method from the aspects of service quality and privacy protection.The experimental results and safety analysis show that compared with the existing methods,this method can enhance the protection degree while ensuring the service quality and reducing the system overhead.