Ad Hoc网络中一种基于救赎机制的信任模型

在Ad Hoc网络中,节点由于自私等恶意原因以及链路错误导致拒绝转发数据,需要采用适当的机制来提高网络的转发性能。提出一种Ad Hoc网络中的信任模型来激励节点间的协作,在此信任模型中,每个节点通过直接监控和其它节点的推荐监控信息共同完成对邻居节点的信任评价,并维护所有邻居节点的信任值,同时在信任值更新和重建中引入信任值衰退和信任救赎机制来体现模型的健壮性。实验表明该信任模型中的信任救赎机制能提高网络可用性,且该模型有效抑制了恶意节点的行为。     

动态信任管理技术研究

现存的P2P信任管理模型难以准确计算节点间的信任值,导致节点在交易过程中无法有效防止节点的恶意攻击.为此,文章提出一种基于推荐信任迭代的信任管理模型TMRTI,通过在间接信任度的计算过程中引入最佳推荐链,更好地计算节点间的推荐信任值,在直接信任度的计算过程中引入历史交易成功率,进而综合评估节点的全局信任值,辅助节点进行交易对象的选择,充分体现了信任值的动态性和主观不确定性.仿真实验结果表明,文章的TMRTI模型较以往模型具有更好的动态适应性和更高的安全性.     

基于加权信任向量的混合结构式P2P网络信任模型

为了解决非结构化P2P网络信任模型的全局信任度的迭代计算和结构化P2P网络信任模型的网络规模不易扩展性的缺点,设计了基于加权信任向量的混合结构式P2P网络信任模型(简称W-TPP),它采用信任向量来存储历史经验数据,利用结构化网络中的节点来分布式存取全局信任向量表.并且为了解决历史经验数据的时间有效性问题,提出一种新的节点信任值算法-加权信任向量算法.通过对W-TPP信任模型的性能分析和模拟仿真试验,验证了该信任模型具有可扩展性,降低了查询全局信任度所占的网络资源.     

基于超级节点的P2P信任模型-TSN

提出一种基于超级节点的P2P信任模型TSN.改进了模型节点加入、节点信任值的计算方法.在模型中综合考虑了节点兴趣相似度和节点响应时延等因素,并通过调节直接信任和间接信任的权重,提高了信任值计算的精确性.采取了相应的健壮性策略,增强了模型的抗攻击能力.仿真结果表明,与已有的信任模型相比,该模型具有较高的成功请求率和系统安全性.     

基于超级节点的P2P信任模型的研究

提出基于超级节点的P2P网络信任模型—Super Trust。在Super Trust模型中,对节点的信任值采用了组内直接信任,组内间接信任和组间信任相结合的方式,从而提高了信任值计算的精确性;此外,通过引入超级节点机制,提高了系统交互的成功率。实验结果表明,与基于推荐信任传统模型和RBTrust模型相比,Super Trust具有较高的交易成功率,并且在不同的恶意节点攻击模式下具有较高的成功交易率。     

基于信任评估的Ad Hoc网络关键节点选取算法

Ad Hoc网络中存在关键节点,它们的失效会严重影响网络性能。为了有效地选取重要且可信的关键节点,给出了关键节点的定义,提出了一种基于信任评估的关键节点选取算法。首先利用节点收缩法计算节点的重要度,然后结合D-S证据理论,建立节点的信任评估模型,通过该模型得到节点的客观信任值来判断节点的可信度,最后综合考虑节点的重要度和可信度来选取关键节点。仿真实验结果表明,通过该方法得到的关键节点失效后将造成网络性能急剧下降。     

基于物联网节点行为检测的信任评估方法

针对物联网感知层节点的信任研究未能兼顾信任评估的主客观性且信任传递能耗大的问题,提出一种基于节点行为检测的低能耗信任评估模型。该模型采用事件触发检测与周期性检测相结合的方式,通过直接信任值、统计信任值与推荐信任值3种信任因子计算综合信任值,进而判断节点行为是否发生异常。仿真实验结果表明本方法有效兼顾信任评估的主客观性,同时可以快速规避恶意节点并降低信任传递能耗。     

使用深度学习的社交物联网信任预测模型

社交物联网是社交网络概念在物联网中整合后兴起的一个蓬勃发展的研究领域。提出了一种适用于社交物联网网络的改进型节点级信任模型,并通过与其他信任模型的对比仿真实验证明在恶意节点的攻击下,提出的模型拥有更好的稳定性和适用性,总体波动较小。同时,针对实际社交物联网网络中新加入网络的陌生节点可能遇到的网络延迟影响信任值评估的问题,在改进型节点级信任模型的基础上进一步使用了深度学习模型对其进行信任值预测。仿真证明,使用深度学习预测后模型的系统性能明显优于不使用深度学习的模型,成功交互率提升约1.8%。     

一种新的基于概率统计论的P2P网络信任模型

经典的P2P网络信任模型采用迭代方法由局部信任值推算出全局信任值,每次交易都要引起整个网络节点的迭代运算,导致计算复杂,通信流量大,并且面临交易数据过于稀疏,计算不够精确从而容易陷于恶意节点的共谋、诋毁和睡眠等攻击问题。为了保证数据的稠密性与计算结果的精确性,论文基于概率统计理论,提出了一种新的全局信任模型。该模型根据节点的历史交易情况,运用最大似然估计,假设检验等方法计算出节点的信任度,节点选择与可信度高的节点进行交易。数学分析与仿真实验表明该模型能有效地抵抗恶意节点的攻击,与经典信任模型Eigentrust相比,较大程度地提高了整个P2P网络的成功交易率。     

基于Beta分布的无线传感器网络分簇路由信任管理机制研究

信任管理机制为保障无线传感器网络安全提供了一种有效方案,通过对节点的行为进行评价,建立整个网络的信任管理机制,对判断的恶意节点采用相应的限制措施来保障安全。文章以Beta概率密度分布函数的期望值作为信任值,优化网络分簇路由协议,可在保障网络能量高效利用的同时,有效提升整个网络的安全性。     

Trust evaluation method for clustered wireless sensor networks based on cloud model

Trust Management has been proved to be an effective method to detect malicious nodes and ensure security in Wireless Sensor Networks (WSNs). While, most existing trust management methods are not good at dealing with uncertainty of trust relationship such as randomness, fuzziness in WSNs, which leads to inaccurate trust metric. In this paper, a trust evaluation method for clustered wireless sensor networks based on cloud model is proposed and evaluated, which implements the conversion between qualitative and quantitative of sensor nodes’ trust metrics in order to achieve better trust evaluation. Firstly, the method considers multi-factors including communication factor, message factor and energy factor and builds mathematical model for each trust factor to get factor trust cloud. Secondly, immediate trust cloud is calculated by assigning adjustive weights for each factor trust cloud and combining them. Thirdly, recommendation trust cloud and immediate trust cloud are synthesized according to time sensitive factor in order to get final trust cloud. Furthermore, the final trust cloud of sensor node is converted to trust grade by trust cloud decision-making. Verification Experiments manifest that the proposed method has feasibility and accuracy in the aspect of evaluating sensor nodes’ trust. Moreover, comparison experiments under different attacks show that our method is sensitive to multiple attacks, it outperforms other trust evaluation methods not only in the accuracy of detecting malicious nodes, but also in the tolerance of abnormal conditions.     

Secure Minimum-Energy Multicast Tree Based on Trust Mechanism for Cognitive Radio Networks

In Cognitive Radio (CR) networks, the non-cooperative behavior is an inherent security issue because it is necessary to realize many functions by means of cooperation, such as spectrum sensing. Then it is important to guarantee the support of cooperation between nodes on multicast communication. In this paper, first the multicast model with optimizing energy use is shown, in which consists of many multicast units. By considering the power consumption concerning spectrum sensing and data transmission, the multicast problem with optimizing energy use is translated into a 0-1 integer programming problem. Secondly, the trust values are calculated by the Bayesian theorem in CR networks. In order to improve the stability of trust mechanism, the new trust values are modified by the iterative control criterion. Finally, a secure minimum-energy multicast (SMEM) algorithm is proposed to ensure multicast communication, and the following example is shown to explain it. The simulation and analysis show that the time complexity of our proposed algorithm is polynomial. Moreover, with the increase of destination nodes, the SMEM algorithm is more effective than the distributed Dist-Implement models of adjustable transmission power in energy utilization.     

Self-Recommendation Mechanism in Trust Calculation Among Nodes in WSN

This paper proposes a node self-recommendation mechanism which is effective in trust calculation model in wireless sensor networks (WSNs). The mechanism has little practical sense to cable and wireless network among which nodes’ resources, especially energy, are almost unlimited; furthermore it may bring some security risk to networks on the contrary. But as to WSNs where nodes’ resources are strictly limited, a node can express its intension of participating communication to its neighbors by using the mechanism according to its current running state and predefined strategies. This mechanism is useful to save nodes’ energy, balance network load and prolong network lifetime ultimately. The paper focuses on self-recommendation value expression, calculation and synthesis method. Application method of the mechanism is also discussed. Simulation results show that using trust calculation model cooperatively with self-recommendation mechanism can protect low energy nodes effectively and balance energy consumption among adjacent nodes without weakening malicious node identification function of the trust model.

     

Trust Based Detection and Elimination of Packet Drop Attack in the Mobile Ad-Hoc Networks

The mobile ad hoc network (MANET) is communication network of a mobile node without any prior infrastructure of communication. The network does not have any static support; it dynamically creates the network as per requirement by using available mobile nodes. This network has a challenging security problem. The security issue mainly contains a denial of service attacks like packet drop attack, black-hole attack, gray-hole attack, etc. The mobile ad-hoc network is an open environment so the working is based on mutual trust between mobile nodes. The MANETs are vulnerable to packet drop attack in which packets travel through the different node. The network while communicating, the node drops the packet, but it is not attracting the neighboring nodes to drop the packets. This proposed algorithm works with existing routing protocol. The concept of trusted list is used for secure communication path. The trusted list along with trust values show how many times node was participated in the communication. It differentiates between altruism and selfishness in MANET with the help of energy level of mobile components. The trust and energy models are used for security and for the differentiation between altruism and selfishness respectively.     

Study on Quantum Trust Model Based on Node Trust Evaluation

Trust is the premise and foundation of se-cure communication, no matter in the quantum commu-nication or classical communication. In fact, the existing quantum secure communication protocols and technolo-gies are implicitly related to trust; or assume some trust premise in advance; or create or obtain a trust relationship. This paper studies on quantum trust model using node trust evaluation based on author's own research achieve-ments. We introduce the trust management into quantum communication network to build secure trusted quantum communication network based on evaluating the trust val-ues of nodes, which are used to evaluate the reliability of each user. We put forward a scheme of quantum trust model based on node trust evaluation, and describe the thought and process of trust evaluation in detail based on the principles of quantum entanglement and quantum teleportation. We analyze the feasibility and safety of this scheme, which provides a new thinking and method for es-tablishing a credible secure quantum communication net-work.     

A Hybrid Trust Management Scheme for Wireless Sensor Networks

Wireless sensor network (WSN) consists of wireless small sensor nodes deployed in the terrain for continuous observation of physical or environmental conditions. The data collected from the WSN is used for making decisions. The condition for making critical decision is to assure the trustworthiness of the data generated from sensor nodes. However, the approaches for scoring the sensed data alone is not enough in WSN since there is an interdependency between node and data item. If the overall trust score of the network is based on one trust component, then the network might be misguided. In this work, we propose the hybrid approach to address the issue by assigning the trust score to data items and sensor nodes based on data quality and communication trust respectively. The proposed hybrid trust management scheme (HTMS) detects the data fault with the help of temporal and spatial correlations. The correlation metric and provenance data are used to score the sensed data. The data trust score is utilized for making decision. The communication trust and provenance data are used to evaluate the trust score of intermediate nodes and source node. If the data item is reliable enough to make critical decisions, a reward is given by means of adding trust score to the intermediate nodes and source node. A punishment is given by reducing the trust score of the source and intermediate nodes, if the data item is not reliable enough to make critical decisions. Result shows that the proposed HTMS detects the malicious, faulty, selfish node and untrustworthy data.     

一种基于邻居合作监测的移动Ad hoc网络信任模型

针对自私节点的恶意丢包行为,将节点和其邻居节点所监测的结果结合起来,计算出节点间的信任度,并以此作为路由选择的依据来促进节点间的相互合作,提出一种基于邻居合作监测的移动Adhoc网络信任模型.将信任模型应用于DSR路由协议并在NS2中进行仿真实验,结果表明该信任模型可以有效地缓解自私节点造成的影响,提高了网络的分组投递率.     

基于云模型的无线传感器网络恶意节点识别技术的研究

 无线传感器网络(Wireless Sensor Network,简称WSN)是一种没有基础设施的自组织无线网络.和其它网络一样,WSN需要安全措施来保证网络通信的安全.但是,在无线传感器网络中,基于密码的安全体系不能有效处理来自网络内部的攻击,识别出恶意节点.因此,信任模型被用于无线传感器网络恶意节点识别.在信任模型和云理论的研究基础上,本文构建了一个基于云理论的无线传感器网络信任模型——云信任模型(CTM,Cloud-based Trust Model).实验结果表明,云信任模型能够有效识别恶意节点.