基于相关系数的证据冲突度量方法

当证据高度冲突时,使用D-S组合规则将会得到不合常理的结果,因此,针对证据组合规则的改进成为一个重要的研究课题,在选择合适的证据组合方法之前,确定证据之间是否冲突、冲突程度如何,是个至关重要的问题,为了更加准确地描述证据间的冲突,在分析证据相似度与证据冲突关系的基础上,基于证据相关系数提出了一种新的证据冲突度量方法,理论分析和数值计算表明该方法可有效表示证据之间的冲突程度,相关系数越大,证据间冲突就越小,与其他方法相比,该方法更具全面性和合理性。     

二线制医院呼叫系统的研制

针对医院四线制呼叫系统存在着系统安装、维护困难和无线呼叫系统易受外界干扰的问题,设计了一种基于单片机的二线制呼叫系统。该系统采用有限状态机模型和＂主机查询＂方式,在保证了系统实时性的同时,有效解决了多个分机同时呼叫主机引起冲突的问题。经过临床应用,该系统有效解决了系统安装、维护困难和易受外界干扰的问题。系统实时性强、可靠性高、成本低、安装及使用方便,具有很强的实用价值。     

采用ADS-B的冲突飞机选择算法

提出了一种采用ADS-B的冲突飞机选择算法,满足未来空城复杂交通态势对自由飞行的需求.算法利用ADS-B提供丰富航行信息及高速更新率等特点,在本机周围40 ～100 nmile范围内,选择出有可能发生冲突的飞机.运用Matlab进行功能仿真,并通过Monte-Carlo方法分析其性能,结果表明算法在排除无威胁飞机方面性...     

基于Haenni方法的证据冲突原因分析

由于Dempster组合规则所存在的不足,在组合冲突信息时,会得出与直觉相反的结论。针对Zadeh悖论,Haenni给出了反驳,但冲突产生的原因分析不够全面。鉴于此,本文在他们的基础上,总结出冲突产生的原因,它不仅与传感器受干扰有关,与识别框架不完整也有关系,其中识别框架不完整还包括框架中的元素不详尽,出现了新目标和在两两证据进行推理过程中目标丢失导致的潜在冲突。     

基于证据理论的住宅规划灰色综合评价方法

针对住宅规划评价指标均属于定性指标,提出了基于证据理论的住宅规划灰色综合评价方法。首先,应用序关系分析法确定评价指标的主观权重系数,其次采用灰色评价方法确定住宅规划关于不同灰类的灰色评估系数,最后采用证据理论将多位专家关于住宅规划的灰色评估系数进行集结,降低了评价结果的主观性。实验结果表明,该方法能够有效地对住宅规划进行评价,评价结果与实际情况相符。     

基于动态聚类和合理性差别的证据理论改进算法

针对证据高度冲突时利用DS证据理论直接进行融合会得出不合理结果的问题,文中提出了一种新的证据融合方法.首先,利用动态聚类对证据进行分类,将含证据最多的类称为主类,其余类称为旁类.当旁类的证据个数与总的证据个数的比值小于或等于设定的阈值时,认为上述证据不合理并删除;当比值大于这一阈值时,则保留该旁类中的证据.然后,提出基于重心距离的方法确定剩余证据的可信度,求出期望证据,最后对其迭代融合.计算机仿真结果验证了该方法的有效性.     

Credibility protection for resource sharing and collaborating in non-center network computing environments

Non-center network computing environments have some unique characteristics, such as instability, heterogeneity, autonomy, distribution and openness, which bring serious issues of security and reliability. This article proposes a brand-new credibility protection mechanism for resource sharing and collaboration in non-center network computing environments. First, the three-dimensional hierarchical classified topology （3DHCT） is proposed, which provides a basic framework for realizations of the identity credibility, the behavior credibility and the capability credibility. Next, the agent technology is utilized to construct the credibility protection model. This article also proposes a new comprehensive credibility evaluation algorithm with simple, efficient, quantitative and able to meet the requirements of evaluating behavior credibility and the capability credibility evaluation as well. The Dempster-Shafer theory of evidence and the combination rule are used to achieve the evaluation of the capability credibility. The behavior credibility is evaluated with the current and historical performance of nodes for providers and consumers to realize more accurate prediction. Based on the non-center network computing simulation test platform, simulation is been conducted to test the performance and validity of the proposed algorithms. Experiment and analysis show that the proposed algorithms are suitable for large-scale, dynamic network computing environments, and able to maintain the credibility for networks without relying on central node, make a non-center network gradually evolve into an orderly, stable and reliable computing environment efficiently.     

Fault Diagnosis Based on the Updating Strategy of Interval-Valued Belief Structures

This paper presents the dynamic method for fault diagnosis based on the updating of Interval-valued belief structures （IBSs）. The classical Jeffrey＇s updating rule and the linear updating rule are extended to the framework of IBSs. Both of them are recursively used to generate global diagnosis evidence with the form of Interval basic belief assignment （IBBA） by updating the previous evidence with the incoming evidence. The diagnosis decision can be made by global diagnosis evidence. In the process of evidence updating, the similarity factors of evidence are used to determine switching between the extended Jef frey＇s and linear updating rules, and to calculate the linear combination weights. The diagnosis examples of machine rotor show that the proposed method can provide more reliable and accurate results than the diagnosis methods based on Dempster-Shafer evidence theory.     

基于GPU的分类并行算法的研究与实现

分析了KNN算法在GPU上实现并行计算的可能性,提出了通过使用CUDA实现KNN算法的方案,在研究了GPU对存储访问的机制后,通过设计合理的数据以及对算法的改进,避免存储体冲突的产生,提高了算法的健壮性。研究结果证明该方法在GPU上的并行运算速度明显要快于CPU,有着很好的加速比。     

一种面向多业务恢复的路由优化算法

讨论了在WDM光传送网中完成多业务恢复时所出现的恢复路由资源竞争问题，并提出了一种基于遗传算法的面向多业务恢复的路由优化算法。该算法可以有效地解决恢复路由资源竞争问题，得到全局最优恢复路由。通过对实际网络的计算，验证了该算法在解决多业务恢复路由优化问题上所具备的优势。