摘要：
In recent years,the rapid development of wireless communication technology has promoted the development of locationbased services (LBS),among which the point-of-interest (POI) query is one of the most important applications.A novel privacy preserving method of k-anonymous model is proposed to solve the problem of leaking location privacy during the query process in road network environment.First,the anonymous server uses the set of points of interest to construct the network Voronoi diagram.Then,the whole road network is divided into independent units which are called network Voronoi cell (NVC) without overlapping.Moreover,the anonymous server uses the Hilbert curve to traverse the road network space and sort the points of interest in accordance with Hilbert order.When a user requests a query,the anonymous algorithm selects dispersed k-1 NVCs which have the same query frequency with the NVC that user located in,and then generates dummy locations in the relative road segments corresponding to the user's in each NVC.The reciprocity of the anonymity set can be ensured and the inference attack that traditional k-anonymity can't resist can be avoided through the proposed anonymous algorithm.Finally,the theoretical analysis and experimental results show that the proposed privacy preserving scheme can effectively protect the location privacy.%近年来,无线通信技术的迅猛发展推动了基于位置服务(location-based services,简称LBS)的发展进程而其中,兴趣点(point of interest,简称POI)查询是基于位置服务最重要的应用之一.针对在路网环境下,用户查询过程中位置隐私泄露的问题,提出了位置k匿名隐私保护方法.首先,匿名服务器将兴趣点作为种子节点生成网络Voronoi图,将整个路网划分为相互独立且不重叠的网络Voronoi单元(network Voronoi cell,简称NVC).其次,利用Hilbert曲线遍历路网空间,并按照Hilbert顺序,对路网上所有的兴趣点进行排序.当用户发起查询时,提出的匿名算法通过查找与用户所在NVC的查询频率相同且位置分散的k-1个NVC,并根据用户的相对位置在NVC内生成匿名位置,从而保证了生成的匿名集中位置之间的相互性,克服了传统k-匿名不能抵御推断攻击的缺陷.理论分析和实验结果表明,所提出的隐私保护方案能够有效地保护用户位置隐私.
