利用分散掩飾的方法來保護使用者的位置隱私

Abstract

隨著定位系統的發展，LBS服務已經成為一種非常重要的應用。而如何保護個人的位置資訊也因此成為一個非常重要的課題。在這篇論文內，我們提出了一個新的觀念來解釋所謂的K匿名來保護使用者的位置隱私。我們同時提出了一個叫做分散掩飾的演算法，以達到更高的K匿名層級和獲得較少的答案候選給LBS的使用者。我們允許使用行動裝置的使用者送出要求LBS服務的查詢，而不需要洩露出他們的精確位置，而且需要很少來自LBS提供者的支援。這個掩飾演算法不僅可以達到高K匿名層級，而且可以獲得較少的答案量。我們實驗也顯露出分散掩飾的演算法可以保護使用者位置隱私保護的很好。

With the progress of mobile devices and positioning technologies, the location-based services (LBSs) have emerged as one of the killer applications. And preserving personal location information becomes a very important issue for the LBS users. In this paper, we propose a new concept of k-anonymity to preserve users' location privacies. We introduce a novel cloaking algorithm, called Split-Cloak, to achieve higher k-anonymity level and obtain fewer candidate answers for the LBS clients. Our model enables the mobile users to send their location-based queries without revealing their exact locations and just need few support from LBS providers. The cloaking algorithm can achieve not only high k-anonymity level but also obtain smaller answer size. Our experiments show that the Split-Cloak algorithm can preserve users' location privacy well.