Verifiable and Privacy-preserving Check-ins for Geo-social Networks

Abstract

Enabling efficient and accurate location verification is a desirable technique for the proliferation of many “check-in” based location sharing services. Existing solutions either use modified infrastructures, such as WiFi APs and cellular towers, to issue location proofs for users, which suffer from high infrastructure extension cost, or rely on co-located users to issue location proofs for each other, which are vulnerable to collusion attacks. In this paper, we propose a novel location verification approach that is based on the concept of location tag. When a user requests to check-in at certain venue, both the device of the user and of the venue will submit a location tag, which actually is a collection of environmental signals presented at a certain location and time, to the service provider. Based on the similarity of the two location tags, the service provider can verify whether the user is actually at the location as she claimed. Due to the high time- and space- dependence of environmental signals, it is hard for users to forge location tags. On the other hand, to protect privacy of a user from the location-sharing service, periodically changed pseudonyms are used by users when checking-in at venues, and partially blind signatures are employed by the services to issue check-in confirmation messages. Then users will submit the check-in confirmation messages in a batch manner, using their true Identities. Thus, the location-sharing service knows who has checked-in at which venue, but does not know when the user did. As a result, temporal privacy of user check-ins can be achieved. Thorough analysis and experiments are conducted to demonstrate the feasibility, security and efficiency of the proposed approach.