SHARP: Private Proximity Test and Secure Handshake with Cheat-Proof Location Tags

Abstract

A location proximity test service allows mobile users to determine whether they are in close proximity to each other, and has found numerous applications in mobile social networks. Unfortunately, existing solutions usually reveal much of users’ private location information during proximity test. They are also vulnerable to location cheating where an attacker reports false locations to gain advantage. Moreover, the initial trust establishment among unfamiliar users in large scale mobile social networks has been a challenging task. In this paper, we propose a novel scheme that enables a user to perform (1) privacy-preserving proximity test without revealing her actual location to the server or other users not within the proximity, and (2) secure handshake that establishes secure communications among stranger users within the proximity who do not have pre-shared secret. The proposed scheme is based on a novel concept, i.e. location tags, and we put forward a location tag construction method using environmental signals that provides location unforgeability. Bloom filters are used to represent the location tags efficiently and a fuzzy extractor is exploited to extract shared secrets between matching location tags. Our solution also allows users to tune their desired location privacy level and range of proximity. We conduct extensive analysis and real experiments to demonstrate the feasibility, security, and efficiency of our scheme.