一种路网环境下LBS隐私保护算法

A privacy protection algorithm designed for LBS in road networks

近年来,基于位置服务(LBS)的应用越来越广泛.用户在享受这种位置服务带来的方便快捷之同时,也要承担可能暴露自身隐私信息(如位置等)的风险.目前,很多工作已经在诸如保护用户隐私信息方面取得了重大进展,但大多是集中于欧氏空间下的隐私保护技术,而路网环境下的隐私保护研究相对较少.针对路网环境下用户边权分布不均的问题,提出了基于哑元的边权均衡算法,即在保证匿名集中各路段邻近性的同时,以生成哑元的方式均衡各边边权分布.这样既能最大程度降低查询代价,又能使边权分布不会太分散.最后,通过实验验证了本算法的有效性,同时显示该算法还能有效防止边权分布不均引发的推断攻击.

In recent years, location based services （LBS） have become more and more popular. However, the user＇s privacy information such as his（her） locations is threatened, while the user enjoys the convenience and effectiveness provided by such services. Many different approaches have been raised to protect the user＇s privacy under Euclidean space. But these approaches rarely focus on conditions under road networks. Now, we propose an edge weight balancing algorithm. The main idea of this algorithm is to create anonymous sets to guarantee the proximity of all edges and then to generate dummies to balance edge weight. Finally, the experiment verifies that the algorithm is able to resist inference attacks caused by uneven edge weight distributions.