LAUSR

As one of the most significant factors for mobile users, continuous locations have been considered in associating with destination in location-based services. In this paper, we consider the destination associated with each location in moving and design an association shifting scheme (ASS) to enhance mobile user’s location privacy. Specifically, we first formalize the model of associating destination with continuous locations such that we can quantify the association. Then we evaluate the model of destination associating with some algorithms. To address this association, we design ASS, which consists of three algorithms that termed as the location shifting algorithm (LSA), the destination constraint algorithm (DCA) and the shifted destination association algorithm (SDAA). In LSA, we shift the real location into a location without association to precise destination, based on the properties of mutual information, which preserves the concealment of precise destination. In DCA, we utilize probability indistinguishable which is a feature inherited from differential privacy to constrain the shifting location, and guides this location to a fake destination with similar probabilities. In SDAA, we leverage the correlation coefficient to map the shifted destination, which helps users to choose fake destination far away from the real one. Moreover, we discuss the security properties of our scheme. We have conducted real experiments on real-world data sets. Experimental results show that our scheme out performs existing approaches in effectiveness and efficiency.