基于协作信誉和设备反馈的物联网边缘服务器信任评估算法

针对边缘服务器的安全问题,提出了一种集成了服务器协作信誉以及设备用户反馈的信任评估算法来提高边缘计算上下文的安全性.交互过程中,使用了一种基于客观信息熵理论的融合算法来聚合服务器间的协作信誉,同时采用了部分同态加密算法来防止交互过程中用户数据的泄露.交互结束后,选择高可信的设备节点计算反馈信任,克服了传统机制的恶意反馈...     

物联网中基于相似性计算的传感器搜索

物联网逐渐成为学术界研究的热点领域,无处不在的传感器设备促进了传感器搜索服务的产生。物联网中搜索的强时空性、海量数据的异构性与传感器节点的资源受限性,给物联网搜索引擎高效地查询传感器提出了挑战。该文提出基于传感器定量数值的线性分段拟合相似性(PLSS)搜索算法。PLSS算法通过分段和线性拟合的方法,构建传感器定量数值的相似性计算模型,从而计算传感器的相似度,根据相似度查找最相似的传感器集群。与模糊集(FUZZY)算法和最小二乘法相比,PLSS算法平均查询精度和查询效率较高。与原数据相比,PLSS算法的存储开销至少降低了两个数量级。     

基于拍卖的物联网搜索数据提供机制

不同于互联网搜索采用爬虫等技术来采集数据,物联网搜索获取数据的主要方式之一是数据拥有者主动提供数据,然而自私性及隐私保护等原因导致数据拥有者不愿主动提供数据。这使得如何有效激励数据拥有者提供数据成为促进物联网搜索发展的关键问题之一。由于竞价拍卖能有效激励参与合作,因此将竞价博弈引入到物联网搜索中,为物联网搜索中的数据采集提供方法。具体地,提出了面向性价比最优的多属性逆向拍卖激励机制,证明了最优数据质量独立于自身报价、其他竞拍者的数据质量及其报价。     

面向物联网的轻量级电子标签安全搜索系统

目前物联网正快速地应用于各个行业,在这些应用中都需要系统能够快速安全地搜索特定的电子标签。在实际应用中,由于物联网中得到大规模应用的标签一般均为低成本的标签,面向物联网的轻量级的电子标签搜索系统所作的研究及其成果可以应用到这些实际中,为物联网中涉及电子标签搜索的需求提供安全、高效的解决方法。     

物联网中物理实体搜索服务的研究

物联网的远景规划是实现物理世界和信息世界的无缝整合,物理实体的搜索服务是物联网发展路线上的关键环节之一.用户利用物理实体的搜索服务从海量动态的物理实体中寻找匹配其需求的物理实体.物理实体除具有静态属性外,还具有实时变化的动态属性,此外还与语境信息紧密相关.本文首先提出一个元数据模型来刻画物理实体及相关的资源和实体服务；然后提出用于衡量物理实体搜索服务的特征,并利用这些特征对具有代表性的搜索服务进行总结、分析和对比,指出它们的优缺点:最后对物理实体搜索服务未来可能的研究策略与突破方向进行展望.     

物联网中适用于内容搜索的实体状态匹配预测方法

对实体匹配用户内容搜索的状态进行预测可显著提高物联网搜索的效率,降低搜索过程的通信开销。该文提出等时距与周期内实体状态预测方法,估计实体在用户查询时刻的状态;设计了适用于内容搜索的有序验证方法,依据实体匹配用户查询内容的概率对实体进行排序验证,以保证用户搜索结果的可靠性。结果表明,所提实体状态预测方法具有较高的精度,结合所提预测方法与匹配验证方法的搜索机制具有较低的通信开销。     

基于组合双向拍卖的物联网搜索任务分配机制

如何合理地分配搜索任务,进而激励用户加入到搜索中是物联网搜索亟需解决的关键问题。针对物联网中数据实效性强的特点,结合物联网搜索中用户的高异构性和动态性,提出一种基于组合双向拍卖的搜索任务分配模型,从市场供求关系的角度描述了搜索发起者、搜索参与者和搜索引擎之间的关系。首先引入了竞价价值的概念,提出了一种基于贪心策略的启发式算法确定竞拍成功的用户集合,然后提出一种基于临界价格的定价算法,确保用户的竞价反映了其真实估价。理论分析及实验结果证明所提任务分配机制在保证激励相容性、合理性的基础上,有效提高了物联网搜索引擎的效率。     

相似度自适应估计的物联网实体高效搜索方法

针对现有相似实体搜索方法缺乏对于观测序列长度的自适应性,且搜索过程数据存储开销过大,搜索结果准确性较低的问题,该文提出相似度自适应估计的物联网实体高效搜索方法(SAEES)。首先,设计了轻量级观测序列分段表示方法,对传感器采集的实体原始观测序列进行轻量级分段压缩表示,以降低实体观测序列的存储开销。然后,提出了观测序列相似度自适应估计方法,实现对不同观测序列长度的实体相似性的准确估计。最后,设计了高效的相似实体搜索匹配方法,依据所估计的实体相似度进行实体的准确搜索匹配。仿真结果表明,所提方法可大幅提高相似实体搜索的效率。

     

物联网中数据融合的信誉度模型与仿真

随着信息时代的到来,物联网技术拥有极具潜力的市场发展前景,以该技术为核心的第三次信息技术浪潮逐步拉来序幕.其中,无线传感器网络集能量、储存以及计算等能力被制约着,进行信息数据之间的有机整合是技术发展的趋势.物联网在应用中被多种不良信息威胁,对技术的使用造成了不同程度上的安全隐患.为了实现信息数据融合的可靠和真实,需要建立安全数据融合模型(CBDA),并基于信誉度评价之上,最终的模型仿真结果从根本上证明了模型的真实可靠以及有效性.     

物联网中基于编码的多路径并发实体通用搜索算法

物联网中许多应用都涉及实体搜索定位。由于物联网中实体类型、通信协议、隶属机构、信息存储组织结构的差异导致了物联网被割裂为相互独立,互不协作的子网络,资源得不到有效利用。基于地标系统的实体搜索定位体系结构是一个开放的、多层多级的、分布式的体系结构,可以有效的整合物联网中的海量实体的信息和各种实体搜索服务。在此体系结构的基础上提出的MCOUS算法采用本地搜索与全局搜索并发进行的多路径搜索策略,高效获取物联网中任意类型实体的位置信息。     

相干声与环境声提取方法的主客观评价关联研究

相干声与环境声的提取有助于实现灵活的空间声重放。不同方法的提取效果需要通过主观测听评估,但是主观测听耗时长、效率低,不利于实时调整算法。客观评价与主观测听相关联,通过客观指标反映主观评价,有利于优化算法、提高效率并保证算法评估的可靠性。本文对已有的四种典型提取方法(主成分分析法、最小二乘法、掩蔽法以及环境声相位估计法)进行主客观评估,其中对比了不同方法提取成分的提取误差和通道间相关值两个客观指标,并将提取成分用于双耳渲染对音质和声像宽度进行主观测听。主客观评估结果表明,提取成分越精确,在双耳渲染中可得到越好的音质;提取的环境声的通道间去相关性越强,在双耳渲染中声像宽度越宽。     

利用主客观集成赋权法的多目标威胁评估

在利用逼近于理想解的排序法(Technique for Order Preference by Similarity to an Ideal Solution,TOPSIS)进行多目标威胁评估时,针对如何获取合理的目标威胁评估因子的权重比,提出了一种主客观集成赋权方法。该方法首先利用主观赋权法和客观赋权法获取两组权重值;然后,通过构造多目标规划模型,将两组权重值进行综合处理,得到更合理的权重值。仿真结果表明,相较于主观赋权法和客观赋权法,所提方法计算出的威胁评估因子的权重值,在用TOPSIS法计算目标威胁评估时,能够得到更加合理、有效的评估结果。     

SOS: Socially omitting selfishness in IoT for smart and connected communities

Smart and connected communities (SCC) is an emerging field of internet of things (IoT), and it has potential applications to improve human life. The improvement may be in terms of preservation, revitalization, livability, and sustainability of a community. The resources of the nodes and devices in the SCC have certain constraints that may not allow the devices and nodes to cooperate to save their resources such as memory, energy, and buffer, or simply maximize their performance. Thus, to stimulate the nodes to avoid selfish behavior, SSC needs a novel and well-organized solution to motivate nodes for cooperation. This article aims to resolve the issue of selfish behaviors in SCC and to encourage the nodes for cooperation. A novel mechanism, socially omitting selfishness (SOS), has been proposed to manage/eradicate selfishness using a socially oriented election process. The election process elects different heads based on weight and cooperation (using Vickrey, Clarke, and Groves model). The election of heads and incentive mechanism encourages the nodes to show participation and behave as highly cooperative members of the community. Furthermore, an extended version of the Dempster Shafer model has been used to discourage the selfish behavior of the participating nodes in the SOS scheme. It uses different monitoring and gateway nodes to efficiently employ the proposed scheme. A mathematical model has been developed for the aforementioned aspects and simulated through NS2 simulation environment to analyze the performance of SOS. The results of the proposed scheme outperform the contemporary schemes in terms of average delivery delay, packet delivery ratio, throughput, and average energy.     

Area,energy, and time assessment for a distributed TPM for distributed trust in IoT clusters

IoT clusters arise from natural human societal clusters such as a house, an airport, and a highway. IoT clusters are heterogeneous with a need for device to device as well as device to user trust. The IoT devices are likely to be thin computing clients. Due to low cost, an individual IoT device is not built to be fault tolerant through redundancy. Hence the trust protocols cannot take the liveness of a device for granted. In fact, the differentiation between a failing device and a malicious device is difficult from the trust protocol perspective. We present a minimal distributed trust layer based on distributed consensus like operations. These distributed primitives are cast in the context of the APIs supported by a trusted platform module (TPM). TPM with its 1024 bit RSA is a significant burden on a thin IoT design. We use RNS based slicing of a TPM where in each slice resides within a single IoT device. The overall TPM functionality is distributed among several IoT devices within a cluster. The VLSI area, energy, and time savings of such a distributed TMP implementation is assessed. A sliced/distributed TPM is better suited for an IoT environment based on its resource needs. We demonstrate over 90% time reduction, over 3% area reduction, and over 90% energy reduction per IoT node in order to support TPM protocols.     

无人机搜索救援系统中机载探测器设计

针对现有的搜索救援系统存在的搜救效率低下和容易发生二次灾害等不足,研究了一种以无线通信技术为核心的无人机搜索救援系统,主要讨论了机载探测器的设计。系统以MSP430微处理器为控制单元,机载探测器通过无线收发模块RFC33A与救援信标机通信,激活信标机并接收信标机获取的遇险人员地理位置信息,通过数传电台XTend将信息实时地传回地面站以及接收地面站的控制指令。实验测试表明,机载探测器对信标机的有效探测距离达到2 km,与地面站的通信距离大约为10 km,工作时间超过3 h,实现了搜索的目的。     

Estimating and eliminating redundant data transfers over the web: a fragment based approach

Redundant data transfers over the Web, can be mainly attributed to the repeated transfers of unchanged data. Web caches and Web proxies are some of the solutions that have been proposed, to deal with the issue of redundant data transfers. In this paper we focus on the efficient estimation and reduction of redundant data transfers over the Web. We first prove that a vast amount of redundant data is transferred in Web pages that are considered to carry fresh data. We show this by following an approach based on Web page fragmentation and manipulation. Web pages are broken down to fragments, based on specific criteria. We then deal with these fragments as independent constructors of the Web page and study their change patterns independently and in the context of the whole Web page. After the fragmentation process, we propose solutions for dealing with redundant data transfers. This paper has been based on our previous work on ‘Web Components’ but also on related work by other researchers. It utilises a proxy based, client/server architecture, and imposes changes to the algorithms executed on the Proxy server and on clients. We show that our proposed solution can considerably reduce the amount of redundant data transferred on the Web. Copyright © 2004 John Wiley & Sons, Ltd.     

An integrated view on monitoring and compensation for dynamic optical networks: from management to physical layer

A vertical perspective, ranging from management and routing to physical layer options, concerning dynamic network monitoring and compensation of impairments (M&C), is given. Feasibility, reliability, and performance improvements on reconfigurable transparent networks are expected to arise from the consolidated assessment of network management and control specifications, as a more accurate evaluation of available M&C techniques. In the network layer, physical parameters aware algorithms are foreseen to pursue reliable network performance. In the physical layer, some new M&C methods were developed and rating of the state-of-the-art reported in literature is given. Optical monitoring implementation and viability is discussed.