基于网格划分的位置隐私保护算法研究
发布时间：2019-04-12 12:03
【摘要】：随着智能移动设备的流行与普及,基于位置的服务应用深入各行各业,改变了人们的生产和生活方式。基于位置的服务应用向用户提供服务的前提是用户提交自己的精确位置。由于位置信息本身具有的敏感特征,位置信息泄露会引起严重的隐私泄露问题,位置隐私问题日益引起人们的重视。位置信息隐私保护方法主要包括假位置、区域匿名、空间加密技术。其中,区域匿名是一种应用广泛,性能高效的方法。区域匿名主要采用位置k匿名模型泛化用户的位置信息达到隐私保护的效果。区域匿名方法目前主要有两大类方法,一类是邻接用户扩展方法,典型算法有:中心伪装算法、最近邻算法和希尔伯特伪装算法、团伪装算法等;另一类是区域扩展方法,典型算法有:四分网格算法、间隔伪装算法、Casper伪装算法、邻接网格扩展算法。Grid-divide算法是一种邻接网格扩展匿名算法。该方法将空间区域依据一定的规律划分成若干个网格,所有网格处于同一层上,用户根据一定策略进行水平或垂直方向上的空间扩展,直至生成的空间区域能满足用户的匿名要求。但是这种算法并不太适合用户分布稀疏或者匿名度要求较高的情况。在Grid-divide算法的基础上提出了一种改进的网格划分位置隐私保护算法。该算法和Grid-divide算法一样采用网格结构划分位置空间,使所有网格处于同一层上,不对其进行层次划分。不同的是进行空间扩展时,采用四周扩展的方式,保证扩展的有效性和扩展粒度较小。此外,引入最大匿名区间参数限制最终形成的匿名区域的大小,借鉴邻接用户扩展的思想采用延迟等待更多用户进入候选匿名区的方式,提高算法的成功率。引入最长可容忍时间参数限制服务延迟的时间,以促进算法更快形成满足条件的较小的匿名区域。采用C语言实现了改进的网格划分空间位置隐私保护算法,使用Brinkhoff基于网络的移动对象生成器生成的数据集进行仿真实验,对比了原始算法和改进算法的匿名时间和匿名效果,并对比了不同参数取值下算法执行情况,证明了改进算法的有效性。
[Abstract]:With the popularity and popularization of intelligent mobile devices, location-based service applications go deep into various industries and change people's production and life style. Location-based service applications provide services to users on the premise that users submit their exact location. Because of the sensitive characteristics of location information, location information disclosure will cause serious privacy disclosure, and the location privacy problem has attracted more and more attention. The privacy protection methods of location information mainly include false location, area anonymity and space encryption technology. Among them, area anonymity is a widely used and efficient method. Regional anonymity mainly uses location k anonymity model to generalize user's location information to achieve privacy protection. At present, there are two main methods of region anonymity, one is adjacency user extension, the typical algorithms are: central camouflage algorithm, nearest neighbor algorithm and Hilbert camouflage algorithm, regiment camouflage algorithm and so on. The other is region extension. The typical algorithms are quartile mesh algorithm, interval camouflage algorithm, Casper masquerading algorithm and adjacent mesh extension algorithm. Grid-divide algorithm is a kind of adjacent mesh extended anonymous algorithm. In this method, the spatial region is divided into several meshes according to a certain rule, all grids are on the same layer, and the user expands the space horizontally or vertically according to a certain strategy. Until the generated space area can meet the user's anonymous requirements. However, this algorithm is not suitable for users with sparse distribution or high anonymity requirements. Based on the Grid-divide algorithm, an improved mesh location privacy protection algorithm is proposed. In the same way as the Grid-divide algorithm, the grid structure is used to partition the position space, so that all meshes are on the same layer and are not hierarchical. In order to ensure the efficiency and granularity of the expansion, the four-dimensional expansion is adopted when the spatial expansion is carried out. In addition, the maximum anonymous interval parameter is introduced to limit the size of the final anonymous region, and the extension of adjacent users is used to delay waiting for more users to enter the candidate anonymous area, so as to improve the success rate of the algorithm. The maximum tolerable time parameter is introduced to limit the time of service delay in order to promote the algorithm to form smaller anonymous regions which satisfy the conditions more quickly. In this paper, an improved privacy protection algorithm for spatial location of mesh partition is implemented in C language. The simulation experiment is carried out with the data set generated by Brinkhoff network-based mobile object generator, and the anonymous time and anonymous effect of the original algorithm and the improved algorithm are compared. The performance of the algorithm with different parameters is compared, and the effectiveness of the improved algorithm is proved.
【学位授予单位】：海南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP309
[Abstract]:With the popularity and popularization of intelligent mobile devices, location-based service applications go deep into various industries and change people's production and life style. Location-based service applications provide services to users on the premise that users submit their exact location. Because of the sensitive characteristics of location information, location information disclosure will cause serious privacy disclosure, and the location privacy problem has attracted more and more attention. The privacy protection methods of location information mainly include false location, area anonymity and space encryption technology. Among them, area anonymity is a widely used and efficient method. Regional anonymity mainly uses location k anonymity model to generalize user's location information to achieve privacy protection. At present, there are two main methods of region anonymity, one is adjacency user extension, the typical algorithms are: central camouflage algorithm, nearest neighbor algorithm and Hilbert camouflage algorithm, regiment camouflage algorithm and so on. The other is region extension. The typical algorithms are quartile mesh algorithm, interval camouflage algorithm, Casper masquerading algorithm and adjacent mesh extension algorithm. Grid-divide algorithm is a kind of adjacent mesh extended anonymous algorithm. In this method, the spatial region is divided into several meshes according to a certain rule, all grids are on the same layer, and the user expands the space horizontally or vertically according to a certain strategy. Until the generated space area can meet the user's anonymous requirements. However, this algorithm is not suitable for users with sparse distribution or high anonymity requirements. Based on the Grid-divide algorithm, an improved mesh location privacy protection algorithm is proposed. In the same way as the Grid-divide algorithm, the grid structure is used to partition the position space, so that all meshes are on the same layer and are not hierarchical. In order to ensure the efficiency and granularity of the expansion, the four-dimensional expansion is adopted when the spatial expansion is carried out. In addition, the maximum anonymous interval parameter is introduced to limit the size of the final anonymous region, and the extension of adjacent users is used to delay waiting for more users to enter the candidate anonymous area, so as to improve the success rate of the algorithm. The maximum tolerable time parameter is introduced to limit the time of service delay in order to promote the algorithm to form smaller anonymous regions which satisfy the conditions more quickly. In this paper, an improved privacy protection algorithm for spatial location of mesh partition is implemented in C language. The simulation experiment is carried out with the data set generated by Brinkhoff network-based mobile object generator, and the anonymous time and anonymous effect of the original algorithm and the improved algorithm are compared. The performance of the algorithm with different parameters is compared, and the effectiveness of the improved algorithm is proved.
【学位授予单位】：海南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP309
【参考文献】
相关期刊论文 前10条
1 刘佳;方贤进;康佳;;社交网络中的位置隐私保护研究[J];电脑知识与技术;2014年28期
2 杨珍;钟诚;杜晓静;;浅析大数据环境下的隐私保护问题[J];电子世界;2014年18期
3 韩建民;林瑜;于娟;贾l，
本文编号：2456995
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2456995.html
