An EM‐based adaptive multiple target tracking filter

Tracking targets of interest is one of the major research areas in radar surveillance systems. We formulate the problem as incomplete data estimation and apply EM to the MAP estimate. The resulting filter has a recursive structure analogous to the Kalman filter. The advantage is that the measurement‐update deals with multiple measurements in parallel and the parameter‐update estimates the system parameters on the fly. Experiments tracking separate targets in parallel show that tracking maintenance ratio of the proposed system is better than that of NNF and RMS position error is smaller than that of PDAF. Also, the system parameters are correctly obtained even from incorrect initial values. Copyright © 2001 John Wiley & Sons, Ltd.     

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.     

羰基铁类随机混合吸波材料等效电磁参数的计算

本文为计及多重散射偶极子间的相互作用，引入参量εｈ和μｈ，导得一组公式。它不仅能计算铁氧体类也能计算羰基铁类的随机混合吸波材料的等效电磁参数，均与实验结果吻合良好。     

EM发酵饲料中主要微生物类群的动态变化

对EM发酵饲料中乳酸杆菌、酵母菌、大肠菌群、腐败菌等微生物类群的动态变化进行了研究。结果表明 :饲料经EM发酵后 ,乳酸杆菌大量增加 ,发酵第4天达到最大值(9.5×108CFU·g-1 饲料) ,大肠菌群与腐败菌被明显抑制。并认为细菌总数与酵母菌数量不能作为评价EM发酵饲料质量好坏的指标。     

SAS数据挖掘在冷轧酸洗卷质量缺陷分析中的应用

数据挖掘作为一种发现大量数据中潜在信息的数据分析技术,受到各界的密切关注。SAS数据挖掘技术是众多数据挖掘方法中的佼佼者,它在大型企业中得到很好的应用。本文介绍数据挖掘的背景知识,并利用SAS/EM工具,对该技术在冷轧酸洗卷质量缺陷原因分析作了初步尝试。     

Deconvolution with partially known kernel of nonnegative signals

This paper extends the Bayesian direct deconvolution method to the case in which the convolution kernel depends on few unknown nuisance parameters. Moreover, an acceleration procedure is proposed that drastically reduces the computational burden. Finally, the implementation of the method by means of the fast Fourier transform is fully discussed in the multidimensional case.     

Soil Apparent Electrical Conductivity (EC<Subscript>a</Subscript>) as a Means of Monitoring Changesin Soil Inorganic N on Heterogeneous Morainic Soils in SE Norway During Two Growing Seasons

An efficient method to monitor changes in soil inorganic N content during crop growth would be a useful means to guide N fertilization to ensure high yields and low N losses to the environment. In this study, soil apparent electrical conductivity (EC_a) measured by the widely used conductivity meter EM38 was tested as an indirect measurement of available N in spring barley during two cropping seasons at two sites with morainic loam in SE Norway. The experiment was constructed to maximize soil variation. In spite of the ȁ8noiseȁ9 caused by the soil heterogeneity, concentrations of inorganic N (cN_inorg) or NO₃-N were most strongly correlated with EC_a in both years and at both locations (with one exception). The measurements of EC_a reflected well the temporal variation in inorganic N content (N_inorg), and a ranking of the treatments based on EC_a fitted very well with a ranking based on N_inorg at the first three sampling times after fertilizing. The best subset of sensor variables (i.e. variables which can be measured ȁ8on-the-goȁ9 by sensor techniques in the field) described 27–69% (average 47%) of the variation in topsoil cN_inorg. When expanding the regression models to include pH as well, the degree of explanation increased significantly. In conclusion, the method of using EC_a appears to be quite robust in terms of detecting relative differences in cN_inorg, whereas a determination of absolute levels of cN_inorg with the method is unreliable.     

带隐变量的回归模型EM算法

带有隐变量的回归模型具有非常广泛的应用场合,隐回归模型的参数求解问题依赖于自变量的分布假设。基于自变量的beta分布的假设条件,给出了隐回归模型的EM算法,详细地推导了模型中的参数求解过程,给出了使用牛顿法求解beta分布参数的算法,并提出一个合适的初值选择算法。在模拟数据和真实数据的基础上进行了详细的比较性试验,结果表明,对具有不同分布特征的因变量观察值,EM算法能够有效地求解隐回归模型的参数。     

基于混合PSO的高斯混合模型地形分类

提出了一种基于改进的混合粒子群优化(particle swarm optimization,PSO)算法的高斯混合模型地形分类方法。高斯混合模型的求解通常是使用期望最大化算法(expectation maximization,EM),然而EM算法易陷入局部最优,收敛速度不稳定且对初值敏感。因此引入混合PSO算法,并对其进行了一系列改进。实验结果表明:改进后的算法较其它优化算法提高了全局搜索能力和收敛速度,利用该算法求解高斯混合模型可以提高参数估计的精度,并且在户外场景图像的地形分类实验中所提出的地形分类方法也表现优良。     

某机载雷达发射校正系统的EMC设计

因受载机平台内电子设备电磁干扰的影响,某机载雷达发射校正时存在通道相位漂移的现象,表现为同一通道相位多次测量偏差超过正常幅度值.在阐述发射校正系统工作原理的基础上,深入分析干扰的来源和机理,提出了综合运用屏蔽、接地、滤波等措施解决电磁兼容问题的具体方法,实现了雷达发射校正系统的可靠工作,为该类工程设计提供了参考.