| With the development of mobile internet and the popularization of smart phones,location-based services are closely related to people's lives.The issue of privacy leakage caused by the publish of location when requesting services gradually attracts the public attention.How to protect users' location privacy while obtaining good location services has become an important research content.In the scenario of real-time online location data publishing,given that the existing location protection method based on differential privacy,only considers the current location protection,and assumes that each publish independent,ignoring the association between current location and personal historical locations or friends' locations,which is difficult to guarantee the privacy of users' locations.Based on the above considerations,on the basis of differential privacy,a location privacy protection model based on personal historical locations and friends' locations is proposed.Analyzed the privacy leakage caused by personal historical locations and friends' locations,combined with the user's real location data and the disturbed location data released by the user and friends,designed the historical location memory mechanism and the selection mechanism based on the friends' locations.According to the location privacy protection model based on personal historical locations and friends' locations,an online location protection method based on personal historical locations and friend's locations is developed and implemented.The theoretical analysis and proof of algorithm time,space complexity,data availability and security are given.Finally,experiments on two data sets containing multiple different cities prove that: under the same privacy budget,compared with existing differential privacy-based methods,it can maintain good data availability while reducing location privacy leakage caused by personal historical locations and friend's locations. |
PDF Full Text Request