Research On Location Privacy Protection For Mobile Social Networks

Recently,the proliferations of intelligent terminals,4G and WIFI have given rise to the development and popularization of the mobile internet.Due to the combination of mobile internet and the traditional social networks,the Mobile Social Network Service(MSNS)has emerged.It not only can facilitate users to discover nearby friends at any time and then promote the interaction between friends,but also allows users to mark the place where the incident occurred when they share photos,video.etc,which makes other users more intuitively understand the shared contents of users.However,location-based service needs to collect locations of users,which exists serious privacy threatens.The location not only contains massive personal privacy information,but also can infer other sensitive information,such as interests,hobbies and health conditions.The untrusted service providers may disclose users' location privacy for commercial purposes.Only relying on existing third-party monitoring mechanism cannot simply ensure the absolute security of users' information.Our contributions in this paper are outlined as follows.Firstly,in order to solve the problem of traditional proximity detection methods which exists the risk of leaking the user's location privacy,we propose a novel privacy-preserving proximity detection method.Specifically,instead of using users'locations to find their neighbors,the proposed method takes beacon signals as a reference of neighbor discovery.More specifically,users can find their neighbors by determining whether users' nearby reference lists have a common item.The method can complete proximity detection without the location of the user,which protects location privacy of the user.Secondly,for overcoming the shortcomings of most existing privacy-preserving proximity detection methods,which cannot support one-to-many proximity detection or calculate the Euclidean distance between users,we present another privacy-preserving proximity detection method.It uses the Geohash encoding to take the location of each user into a bit string by dichotomy approximation,and then divides this string into two subsets:prefix and suffix.The MSNS calculates the candidate neighbors in the same rectangular region according to the location prefix,and then the third-party server calculates relative distances of candidate neighbors according to the location suffix.The method cuts the location information into two separate parts,which avoids the service provider obtaining the complete location information of the user,and effectively improves the privacy security of the user.Finally,for the problem of location sharing service has caused some serious personal privacy leakage threats,we propose a privacy-preserving location sharing scheme without the trusted cellular tower.The mechanism uses the Bloom Filter to separate the Location Server(LS)and MSNS.It can cut off the association between the user identity and location information,and then hides location of the user.Moreover,the mechanism adopts k-Trajectory mechanism,which generates k-1 indistinguishable false locations for each user to prevent reasoning attack in particular locations.The mechanism can achieve various location sharing services under the guarantee of protecting location privacy of the user.