Estimation for the number of components in a mixture model using stepwise split-and-merge EM algorithm

The main difficulty with EM algorithm for mixture model concerns the number of components, say g. This is the question of model selection, and the EM algorithm itself could not estimate g. On the contrary, the algorithm requires g to be specified before the remaining parameters can be estimated. To solve this problem, a new algorithm, which is called stepwise split-and-merge EM (SSMEM) algorithm, is proposed. The SSMEM algorithm alternately splits and merges components, estimating g and other parameters of components simultaneously. Also, two novel criteria are introduced to efficiently select the components for split or merge. Experimental results on simulated and real data demonstrate the effectivity of the proposed algorithm.     

Sol-Gel Preparation of LaNiO3 and Its Physical and Chemical Properties

LaNiO3 was synthesized by sol-gel method in which lanthanum nitrate and nickel nitrate were used as start materials and citric acid was used as complex for gel formation.The precursor was dried and subsequently heated at elevated temperature to form the desired product.XRD analysis shows that pure LaNiO3 was synthesized.Electrical conductivity and electrochemical performance of the material were tested.The electrical conductivity decreases from 34.5that there are current peaks in the curve, which is the evidence of the electrochemical activity of LaNiO3.     

Photoluminescence and X-ray Diffraction of Distributed Bragg Reflector

Spectral and structural characteristics of distributed Bragg reflector (DBR) in vertical-cavity surface-emitting lasers were studied with photoluminescence and double- crystal X- ray diffraction measurement. The expected high quality epitaxial DBR structure was verified. In the X- ray double- crystal rocking curves of DBR the zeroth- order peak, the first and second order satellite peaks were measured. Splitting of diffraction peak appeared in the rocking curves was analyzed. The effects of introduced deep energy levels on the structural perfection and optical properties were discussed.     

Distributed Optical Fiber Sensor for Multi-point Temperature Measurement

The distributed optical fiber sensing technology is overviewed, which is based on Raman scattering light theory. Basic operation principle, structure, system characteristics and signal processing are discussed. This structure and method of the signal processing possess of certain spatial resolution, hence will ensure the practicability of system.     

一种改善Turbo乘积编码OFDM系统性能的新方法

本文提出一种利用对译码器软信息限幅来改善多径衰落信道中Turbo乘积编码OFDM(TPC-OFDM)系统性能的新方法。通过对不同多径衰落信道中QPSK映射和16QAM映射的TPC-OFDM系统性能的数值仿真，结果表明在10^-5误比特率下，这种新方法比传统的迭代译码大约有6～10dB的改进，对严重多径环境下TPC-OFDM系统的错误平底也有明显的改进。     

Itemset trees for targeted association querying

Association mining techniques search for groups of frequently co-occurring items in a market-basket type of data and turn these groups into business-oriented rules. Previous research has focused predominantly on how to obtain exhaustive lists of such associations. However, users often prefer a quick response to targeted queries. For instance, they may want to learn about the buying habits of customers that frequently purchase cereals and fruits. To expedite the processing of such queries, we propose an approach that converts the market-basket database into an itemset tree. Experiments indicate that the targeted queries are answered in a time that is roughly linear in the number of market baskets, N. Also, the construction of the itemset tree has O(N) space and time requirements. Some useful theoretical properties are proven.     

国外尾矿坝安全与环境研究综述

安全与环境一直是国外尾矿坝研究的重点。笔者综述了自1990年以来国外涉及尾矿坝安全与环境方面有代表性的文献 ,内容包括尾矿坝溃坝、溃坝理论研究、安全管理、环境污染、污染监测、污染治理、污染理论研究及生态复垦 ,这些对提高我国尾矿坝管理、工程技术水平都有较好的参考价值     

垂直管吸收器内泡式吸收过程的数值研究

建立垂直管吸收器管内泡式吸收过程中传热传质的数学物理模型，对其泡式吸收过程进行数值研究，获得泡式吸收方式的一些传热传质特性，为吸收器的优化设计提供一定理论指导。     

Extracting Modes of Constrained Structure with Elastic Supports from Free Test Data

The dynamic flexibility (DF) and improved dynamic flexibility (IDF) methods can be applied to extract constrained structural modes from free-free modal test data. The residual flexibility method is also good for boundaries of constrained structure with rigid supports. Under elastic support boundary conditions both the DF and residual flexibility methods cannot produce accurate results. This paper expands the previously published IDF method to become a more general approach for structural design engineers. A new method called the general dynamic flexibility (GDF) method has been developed in this paper to extract constrained structural modes from free test data. The GDF method can always be applied as follows: (1) when the boundary support stiffness is very stiff, the GDF method can accurately obtain results from rigid support as by Liu et al. in 2001 and Zhang and Wei in 2003; (2) when the support stiffness is soft, the present method can produce results similar to those free-free modal parameters measured by hanging structures using rubber band; (3) when the support stiffness is zero, the method can reproduce true free-free modal parameters; and (4) when the support stiffness has any finite values, the GDF method can also achieve satisfactory results for engineering use. The algorithm included in the GDF method converges rapidly and is numerically stable in the analysis. This feature is very practical for many engineering applications for using the GDF method.     

Approach for Overcoming Numerical Inaccuracy Caused by Load Discontinuity

It was found that the discontinuity at the end of an impulse will lead to numerical inaccuracy as this discontinuity will result in an extra impulse and thus an extra displacement in the time history analysis. In addition, this extra impulse is proportional to the discontinuity value at the end of the impulse and the size of integration time step. To overcome this difficulty, an effective approach is proposed to reduce the extra impulse and hence the extra displacement. In fact, the novel approach proposed in this paper is to perform a single small time step immediately upon the termination of applied impulse, whereas other time steps can be conducted by using the step size determined from accuracy consideration in period. The feasibility of this approach is analytically explored. Further, analytical results are confirmed by numerical examples. Numerical studies also show that this approach can be applied to other step-by-step integration methods. It seems that to slightly complicate the programming of dynamic analysis codes is the only disadvantage of this approach.