无线传感器网络中的分布式节点定位方法

提出一种基于流形学习的分布式Hessian局部线性嵌入（DHLLE）定位方法,给出了基于流形学习算法的定位框架.DHLLE方法采用同情最邻近算法来选择节点邻居列表,并应用Hessian局部线性嵌入（HLLE）算法获取传感器网络节点的局部映射,再通过对局部映射合并获得所有节点的全局映射,最后通过对参考节点进行坐标匹配以取得所有节点的全局坐标.仿真结果表明,DHLLE方法能够快速、准确地对节点进行定位,且复杂度低,节点能耗小,其性能超过了分布式加权多维定标等算法.     

基于最小二乘支持向量机和证据理论的交通数据融合

针对基于浮动车辆数据(floating car data,FCD)的城市道路交通信息采集系统存在的问题,提出一种基于最小二乘支持向量机(LS-SVM)和证据理论的数据融合方法,通过融合地感线圈采集的交通流量信息,提高FCD系统交通速度信息采集的准确性.利用LS-SVM回归得到速度-流量关系曲线的临界速度参数,再根据历史数据库用统计方法计算出流量-速度关联规则的可信度矩阵,在得到这些经验知识的基础上,定义了两种证据源的基本概率分配函数.最后,通过D-S证据理论对两种证据源进行数据融合,获得融合后的速度信息.实地跑车实验结果论证了融合算法的有效性和可靠性.     

关于DirichletL-函数的四次加权均值分布

本文利用广义Kloostermann和的定义、Gauss和的定义、特征和估计、三角和估计及其解析方法讨论了Dirichlet卜函数的四次加权均值分布问题，得到两个有趣的加权均值分布公式．     

Data fusion based on fuzzy measures

Choquet integral based on fuzzy measure is a very popular data fusion approach. A major problem in applying the Choquet integral is how to determine a large number of fuzzy measures as the number of attributes increases. The λ-fuzzy measure proposed by Sugeno is a powerful method to resolve this problem. However, the modeling ability of the λ-fuzzy measure is too limited to satisfy actual requirements. In this paper, an extended λ-fuzzy measure is proposed using Shapley value index, and the limitation of the λ-fuzzy measure is significantly overcome under little additional computational loads. The extended fuzzy measure has stronger modeling power than the λ-fuzzy measure, straightforwardly representing interaction among attributes. We apply the extended fuzzy measure to an artificial data set and a real dataset in an iron-steel plant. The results verify the usefulness of the extended fuzzy measure compared with other main existing methods.     

分布式环境下全局模糊频繁项目集的快速挖掘

为了解决模糊关联规则挖掘算法需要用户事先给定模糊集和相应隶属度函数的问题，提出基于分布式聚类自动生成模糊集及隶属度函数的算法GFAM．该算法利用分布式K—Means聚类算法对每个数值型属性进行聚类，求得聚类中心，由此构造全局模糊集，定义全局隶属度函数．DFAR算法根据构造的全局模糊集及隶属度函数进行分布式模糊频繁项目集的快速挖掘，采用全局一局部站点模式，其中包括局部模糊频繁项目集产生算法FLF和全局模糊频繁项目集产生算法FGF．实验结果表明，该算法能准确地生成全局模糊频繁项目集，在求解全局模糊频繁项目集过程中，传送局部模糊候选项目集支持数的通信量为O（n），提高了算法的挖掘效率．     

Data fusion based on fuzzy measures

Choquet integral based on fuzzy measure is a very popular data fusion approach.A major problem in applying the Cho- quet integral is how to determine a large number of fuzzy measures as the number of attributes increases.Theλ-fuzzy measure proposed by Sngeno is a powerful method to resolve this problem.However,the modeling ability of theλ-fuzzy measure is too limited to satisfy actual requirements.In this paper,an extendedλ-fuzzy measure is proposed using Shapley value index,and the limitation of theλ-fuzzy measure is significantly overcome under little additional computational loads.The extended fuzzy measure has stronger modeling power than theλ- fuzzy measure,straightforwardly representing interaction among attributes.We apply the extended fuzzy measure to an artificial data set and a real dataset in an iron-steel plant.The results verify the usefulness of the extended fuzzy measure compared with other main existing methods.     

Enhancement of monacolin K production via intergeneric protoplast fusion between Aspergillus terreus and Monascus anka

Intergenric protoplast fusion between Aspergillus terreus CA99 and Monascus anka M-3, the high and low producers of monacolin K respectively,was performed for enhancement of monacolin K production. The 24-hour-old mycelia of A. terreus CA99 and M. anka M-3 were treated with 0.5% lywallzyme, 0.3% snailase and 0.3% cellulase at 34℃ for 5 h and at 30℃ for 3.5 h, and their protoplasts formation reached 1.76×107/mL and 1.68×107/mL respectively. Parental protoplasts were irradiated with a 30 W UV light away from 30 cm for 3 min and then mixed. The mixture was incubated with 30% PEG 6000 for 15 min. The reviving fusants were isolated on the regeneration plates. Of the 363 fusants isolated, over 100 showed enhanced monacolin K production compared with the parental strain M. anka M-3. Ten of them produced monacolin K about 1.6-fold of that M.anka M-3 does and the monacolin K titer of two fusants (F49 and F104) increased by about 1-fold. The monacolin K yields of F49 and F104 were 460 μg/mL and 457 μg/mL respectively. In optimized fermentation medium, the monacolin K titer of F49 reached 1216 μg/mL.     

用真空温压技术制备纳米金属铜块体材料

采用自悬浮-真空温压法制备纳米金属Cu块体材料。通过X射线衍射(XRD)、扫描电镜(SEM)、透射电镜(TEM)、正电子湮没(PAS)实验、HMV-2型显微硬度计和用Arechimedes原理等测试手段研究压制压力、保压时间、温度等工艺参数对样品密度的影响;考察其热稳定性和微结构。研究结果表明:纳米金属Cu块体材料的密度随压制压力、保压时间和温度的增加而提高,最高相对密度为96.15%;纳米金属Cu块体材料具有较好的热稳定性;在1.5 GPa和200℃压制的样品的平均晶粒度为25.3 nm,微应变为0.047%。     

环流循环除尘系统分离柱内旋风流场分析

由柱坐标系下的Navier-Stokes方程出发,建立了环流循环除尘系统分离柱内流速场新的数学模型,利用具有成熟使用经验的加权残值法对流场进行求解,并与实验结果进行了比较。结果表示,所采用的数学模型和计算方法是可行的。     

中间件技术研究

文章概述了分布式计算中的中间件技术的现状,比较了分布式计算技术和传统计算技术,并简要分析了中间件技术的发展趋势。