一种新的多用户位置隐私保护方案

位置服务中许多应用都需要多用户进行位置共享，但传统的刚性隐私保护策略迫使用户不得不共享自身的精确位置。为满足用户个性化的隐私保护需求，本文提出了一种新的多用户位置共享隐私保护方案。方案利用两种位置转换模型保证用户能够灵活地自定义其不同精度的共享位置，再通过基于中国剩余定理的多秘密共享机制将不同精度的位置数据打包转换，只需一次份额分发就可实现不同共享对象恢复出不同精度的位置信息。方案安全性高，不依赖于位置服务器的可信度，能够抵抗单个位置服务器攻击、多个位置服务器合谋攻击以及多个用户之间的合谋攻击，在不可信环境下实现了多用户位置共享的弹性隐私保护。仿真实验分析进一步表明，该方案与同类方案相比更为高效，性能平稳。当参与存储管理份额的位置服务器数量增加或需恢复更高精度位置信息时，计算通信开销不会剧增，在资源受限的网络环境中同样适用。

A New Location Privacy Protection Scheme for Multi-users

Many applications in location-based services require multiple users to share location, but the traditional rigid policy of privacy protection forces users to share and expose their precise locations. In order to meet the user''s personalized needs for privacy protection, a new multi-user location sharing privacy protection scheme is proposed in this paper. Two kinds of location conversion models are used in our solution to ensure that users can flexibly customize their shared locations with different precision levels. Then all the location data with different precisions is packaged and converted through a multi-secret sharing mechanism based on the Chinese Remaining Theorem so that only one time shares-distribution is needed to realize different shared objects recover location information with corresponding different accuracy. It is proved that the flexible location privacy protection of multi-user location sharing is realized in an untrusted net environment by our solution. It is highly secure which does not depend on the trustability of the location provider and can resist single location server attack, collusion attacks of multiple location servers and multiple users. Furthermore, compared with similar solutions, the simulation experiments show that ours is more efficient and stable in performance. When the number of location servers participating in the shares storage increases or higher-precision location information needs to be reconstructed, the overheads of computation and communication will not increase dramatically. Thus, it is very applicable in the resource-constrained network environment, too.