Personalised Location Privacy Protection based on Grid Division for Crowd-Sensing

Abstract

The existing generalization-based location privacy protection scheme of Crowd-Sensing cannot balance user privacy needs and data quality, and uses a uniform policy to protect all locations of all users. To address this problem, this paper proposes a personalized privacy protection scheme, which first calculates the sensitivity of each location based on location entropy, access frequency and other features, and then encodes the location using Geo-hash code. The prefixes of different lengths are chosen according to the different sensitivities to achieve different strengths of privacy protection. Finally, a reasonable gridding algorithm is designed so that the data quality does not degrade as the protection strength increases, thus achieving the goal of improving data quality while protecting the user's location privacy. Finally, the effectiveness of the proposed algorithm is further verified through experiments.