面向云计算应用层演化的隐私保护方法研究

云计算应用层中的组合服务具有演化属性,因此,隐私数据在服务组合过程中,用户的隐私数据可能会因为服务或服务流程的演化而暴露。根据服务演化的特征,以描述逻辑为基础,提出了一种面向云计算应用层演化的隐私保护方法。对隐私协议进行形式化描述;根据服务的演化特征,对服务的演化进行监控,保证满足用户的隐私需求;利用实例研究证明该方法的正确性与可行性。     

面向连续查询的敏感语义位置隐私保护方案

针对连续查询位置服务中构造匿名区域未考虑语义位置信息导致敏感隐私泄露问题，通过设计[(K,θ)]-隐私模型，提出一种路网环境下面向连续查询的敏感语义位置隐私保护方案。该方案利用Voronoi图将城市路网预先划分为独立的Voronoi单元，依据用户的移动路径和移动速度，选择具有相似特性的其他[K-1]个用户，构建匿名用户集；利用匿名用户集用户设定的敏感语义位置类型和语义安全阈值，以及用户所处语义位置的Voronoi单元，构建满足[(K,θ)]-隐私模型的语义安全匿名区域，可以同时防止连续查询追踪攻击和语义推断攻击。实验结果表明，与SCPA算法相比，该方案在隐私保护程度上提升约15%，系统开销上降低约20%。     

大数据时代大学生个人隐私保护问题研究

伴随着互联网+的风生水起,信息量暴增的大数据时代应运而来,"网络原住民"大学生不可避免、理所当然地处于大数据洪流之中,但也为其个人隐私保护带来了极大挑战。文章针对大数据带来的安全隐私问题,指出大学生个人隐私泄露的可能原因,提出可以采取的个人隐私保护措施,给出大数据时代隐私保护几种技术,期望能给相关人士提供有效参考。     

粒子群属性聚类的位置隐私保护

针对基于位置服务中连续查询情况下，用户自身属性信息很容易被攻击者获取，并通过关联获得用户位置隐私的情况，提出了一种利用粒子群聚类加速相似属性用户寻找，并由相似属性匿名实现用户位置泛化的隐私保护方法。该方法利用位置隐私保护中常用的可信中心服务器，通过对发送到中心服务器中的查询信息进行粒子群属性聚类，在聚类的过程中加速相似属性用户的寻找过程，由相似属性用户完成位置泛化，以此实现位置隐私保护。实验结果证明，这种基于粒子群属性聚类的隐私保护方法具有高于同类算法的隐私保护能力，以及更快的计算处理速度。     

WSNs中规避攻击者的源位置隐私保护路由协议

为了保证无线传感器网络（Wireless Sensor Networks，WSNs）中的源节点位置隐私安全，同时实现安全性能和网络能耗的均衡，提出了WSNs中规避攻击者的源节点位置隐私路由协议，该协议假设节点具有检测攻击者的能力，通过发出危险警告消息使路由路径上的节点采取路由改变策略，使攻击者无法回溯到源节点，延长了源节点保持位置隐私的安全时间。理论分析和仿真实验表明，该协议在消耗较少的通信开销的情况下保证了源节点位置的高度隐私。     

A novel proximity graph: Circular neighborhood cell graph for histopathological tissue image analyzing

The cell is the smallest unit of living beings, which has structural and functional properties. Almost all cell behaviors are regulated by various intracellular reactions initiated by the signaling. The signaling and the distance between cells influence each other. Thus, cell-location-based modeling and analyzing of histopathological tissues provide important information to the expert. In literature, methods such as distance-based threshold, K-Nearest Neighbor, Voronoi graphs, Delaunay triangulation, and colored graph have been used. However, circular neighborhood relationships of cells have not been considered by any CAD system so far despite of its crucial impact. Thus, we developed the circular neighborhood cell-graph. Histopathological images of liver were classified by using features extracted from T-Distance, K-Nearest Neighbor, Voronoi, Delaunay, and the proposed cell-graph. Then, the classification performances of the methods were compared. Experimental results show that liver tissue images can be classified with accuracy of 95.7% by using the features provided by the proposed cell-graph model.     

Photo-crosslinking: An Emerging Chemical Tool for Investigating Molecular Networks in Live Cells

Studying protein–protein interactions (PPIs) is useful for understanding cellular functions and mechanisms. Evaluating these PPIs under conditions as similar as possible to native conditions can be achieved using photo-crosslinking methods because of their on-demand ability to generate reactive species in situ by irradiation with UV light. Various fusion tag, metabolic incorporation, and amber codon suppression approaches using various crosslinkers containing aryl azide, benzophenone, and diazirines have been applied in live cells. Mass spectrometry and immunological techniques are used to identify crosslinked proteins based on their capture transient and context-dependent interactions. Herein we discuss various incorporation methods and crosslinkers that have been used for interactome mapping in live cells.     

增强现实中基于位置安全性的LBS位置隐私保护方法

为了解决基于位置的服务(LBS)和增强现实(AR)技术快速发展带来的用户位置隐私泄露的隐患,分析了现有的位置隐私保护方法的优缺点,提出基于位置安全性的位置隐私保护方法。将区域安全度和伪装区域引入该方法中,将提示某区域是否需要保护这一度量标准定义为区域安全度,非安全区域(即需要给予保护的区域)的区域安全度设置为1,安全区域(即不需要保护的区域)设置为0,通过扩大区域安全度和识别等级来计算位置安全度。实验结果表明,该方法与未引入位置安全性的方法相比降低了平均定位误差,提高了平均安全性,从而有效地保护了用户的位置隐私,提高了LBS的服务质量。     

Weak Delocalization in Graphene on a Ferromagnetic Insulating Film

Graphene has been predicted to develop a magnetic moment by proximity effect when placed on a ferromagnetic film, a promise that could open exciting possibilities in the fields of spintronics and magnetic data recording. In this work, the interplay between the magnetoresistance of graphene and the magnetization of an underlying ferromagnetic insulating film is studied in detail. A clear correlation between both magnitudes is observed but through a careful modeling of the magnetization and the weak localization measurements, that such correspondence can be explained by the effects of the magnetic stray fields arising from the ferromagnetic insulator is found. The results emphasize the complexity arising at the interface between magnetic and 2D materials.     

Low complexity quantum private queries protocol

Private information retrieval(PIR) is an important privacy protection issue of secure multi-party computation, but the PIR protocols based on classical cryptography are vulnerable because of new technologies,such as quantum computing and cloud computing. The quantum private queries(QPQ) protocols available, however, has a high complexity and is inefficient in the face of large database. This paper, based on the QKD technology which is mature now, proposes a novel QPQ protocol utilizing the key dilution and auxiliary parameter. Only N quits are required to be sent in the quantum channel to generate the raw key, then the straight k bits in the raw key are added bitwise to dilute the raw key, and a final key is consequently obtained to encrypt the database. By flexible adjusting of auxiliary parameters θ and k, privacy is secured and the query success ratio is improved. Feasibility and performance analyses indicate that the protocol has a high success ratio in first-trial query and is easy to implement, and that the communication complexity of O(N) is achieved.