基于Client/Server应用层包跟踪技术的CSPA软件开发和仿真

本文介绍了基于跟踪 Client和 Server端的应用层包 ,开发了得到响应和请求时间及包大小分布的 PDF的应用软件 CSPA,通过对 PDF的分析而建立通用 PDF数学模型并用于网络仿真的应用过程     

基于co-occurrence相似度的聚类集成方法

首先提出了一种基于属性值的co-occurrence相似度概念,通过对其进一步的研究,提出了3个等价性表述;然后对属性值之间的co-occurrence相似度进行引申,给出了数据对象之间co-occurrence相似度的定义,并将其成功应用到聚类集成方法中。利用co-occurrence相似度在计算某个初始聚类结果中数据对象之间的相似度时,充分考虑了其他初始聚类结果和该初始聚类结果之间的相互影响和联系。实验表明, 基于co-occurrence相似度的聚类集成（CSCE）方法能有效识别数据之间的细微结构,有助于提高聚类集成的效果。     

基于聚类集成的用户负荷模式识别

为提高错峰管理中用户负荷模式识别的可靠性与普适性,针对目前单一聚类算法难以解决用电负荷数据的不平衡性以及时序特性等问题,提出一种基于聚类集成技术的用户负荷模型识别方案。利用多种标准化方法以及经遴选的聚类算法生成多样化的聚类成员,通过将所有聚类成员合并构造共识矩阵并进行重构,得到较单一聚类算法更为优越的分群结果。该方案比采用单一的聚类分析得到的用户用电负荷数据分簇结果更稳健可靠,且对数据结构变化的敏感度低、分簇效果更好、泛化能力更强,并在中山市6 500家专变用户的用电负荷模式识别中取得了良好的应用效果。     

Cluster ensemble framework based on the group method of data handling

Cluster ensemble is a powerful method for improving both the robustness and the stability of unsupervised classification solutions. This paper introduced group method of data handling (GMDH) to cluster ensemble, and proposed a new cluster ensemble framework, which named cluster ensemble framework based on the group method of data handling (CE-GMDH). CE-GMDH consists of three components: an initial solution, a transfer function and an external criterion. Several CE-GMDH models can be built according to different types of transfer functions and external criteria. In this study, three novel models were proposed based on different transfer functions: least squares approach, cluster-based similarity partitioning algorithm and semidefinite programming. The performance of CE-GMDH was compared among different transfer functions, and with some state-of-the-art cluster ensemble algorithms and cluster ensemble frameworks on synthetic and real datasets. Experimental results demonstrate that CE-GMDH can improve the performance of cluster ensemble algorithms which used as the transfer functions through its unique modelling process. It also indicates that CE-GMDH achieves a better or comparable result than the other cluster ensemble algorithms and cluster ensemble frameworks.     

Experimental characterization of Silicon Drift Detector for X-ray spectrometry: Comparison with theoretical estimation

Reverse saturation current and the ideality factor (η) are the main parameters that affect the performance of a radiation semiconductor detector in different space environmental conditions. We have measured both of these parameters for the Silicon Drift Detector (SDD) used as a radiation detector in the X-ray spectrometry for space borne applications having the active area of 40 mm² and 109 mm² with 450 μm thick silicon. The measured reverse saturation current is compared with the theoretically estimated values using diode equation for various detector operating temperatures and shown that there is a strong dependence of reverse saturation current with ideality factor. Subsequently, using the reverse saturation current ratio method, the slope ratio for small area to the large area SDD is derived and compared with the theoretical slope ratio obtained using the measured ideality factor. It is shown that the slope ratios closely match with the diode equation of the form which has the ideality factor in both the product and exponential terms for these SDDs. The measured spectral energy resolution is ∼150 eV at 5.9 keV for both small and large area SDDs when operated at −40 °C and −65 °C respectively. The noise performance of the spectrometer is also measured in terms of Equivalent Noise Charge (ENC) for various detector operating temperatures and shown that the value of ENC in rms noise electrons is minimal for the pulse shaping time of 3.3 μs.