Researches On Location Privacy Protection Based On K-anonymous

With the rapid development of various mobile devices and positioning technologies,LBS(Location-Based Services)have been widely adopted.By using LBS services,people can easily access corresponding services on locations such as bars,hospitals,nearby friends,etc.Although LBS bring conveniences,it maintains a serious threat to user's privacy because users are required to send his/her location information to the server.If attackers obtain the data,they can figure out sensitive information.For example,he can predict a user's physical condition once the user stays in a hospital.Therefore,the privacy of people who rely on LBS should be protected by certain measures.At present,a variety of protection mechanisms have been proposed,most of which are implemented by TTP(trusted third party)of k-anonymous.This thesis first summarizes the various models of the current location privacy,and analyzes the advantages and disadvantages of the existing location privacy protection technology.Then,based on the central server model,the problems of service quality and privacy protection,k value selection and mass equal query request are studied.Specific content are as follows:First,a prevalent location privacy method based on k-Anonymizing Spatial Region(k-ASR)is proposed to achieve privacy protection by sacrificing quality of service(QoS)while users cannot obtain accurate query results.Our proposal brings two significant features: the optimal k value for the current user is calculated according to the user's environment and social attributes;instead of forming an Anonymizing Spatial Region(ASR),the trusted third party(TTP)generates a fictitious trajectory which contains k location nodes based on the properties of the Markov chain.Meanwhile,LocationBased Services(LBS)requests are conducted with the form of trajectory.Second,a location privacy protection method based on k-anonymous for massive homogeneo us request is proposed.This method no longer randomly searches closest k mobile locations of users,but enables quantities of requests clustered and formed into ASR to satisfy the privacy needs of group users according to the similarity of location,which can massively reduces calculation.To avoid impacting mobile users' experience,a time threshold is introduced in the anonymity process in ca se that the user has to wait for a long time when located in remote areas without data.At the same time,a threshold of anonymous radius is introduced to ensure effective protection of user's location when radius gets extremely low.Finally,the effectiveness of our proposed algorithm is indicated by MATLAB simulation.