Using broadcast to protect user privacy in location-based applications

Abstract

Most existing cloaking techniques are based on a three-tier architecture: mobile users, a trusted anonymizer server, and service providers. In this paper, we propose a new three-tier architecture, where the trusted anonymizer server also serves as a broadcast server. The trusted anonymizer server takes a proactive approach. It first groups mobile users into clusters. Then for each cluster, it fetches the most popular query results from service providers. The query results are then broadcast through an air channel to reach all mobile users. As a result, to get a query answered, a mobile user can first tune into the air channel to determine if the query result is available. If not, the mobile user just sends a traditional location-based query to the trusted anonymizer server. Other than the novel new three-tier architecture, we also propose two cell-based clustering algorithms, and a broadcast index to facilitate the download of query results. The proposed techniques are compared using simulation against the improved Interval Cloak technique under the traditional three-tier architecture. The extensive results show that our system is better in reducing communication cost and protecting user privacy.