基于贝叶斯法和蒙特卡洛仿真的威布尔型装备器材需求预测

为了解决装备器材历史需求数据少、需求规律不明确的问题,提出一种基于贝叶斯法和蒙特卡洛仿真的威布尔型装备器材需求预测方法。针对威布尔分布尺度参数未知以及形状和尺度参数均未知两种情况,分别基于贝叶斯方法通过解析求解和数值模拟的方式进行了参数估计,并引入柯尔莫哥洛夫-斯米尔诺夫检验法对寿命分布模型进行拟合优度检验;综合考虑修复性维修、预防性维修和装备器材的已使用时间,提出了基于蒙特卡洛仿真的部队装备器材年度需求预测方法。算例分析表明：小样本下通过贝叶斯估计得到的寿命分布模型拟合度高,基于仿真的需求预测方法简单、有效。     

DETERMINATION OF THE FAVORED PATTERN IN NEURAL SPIKE TRAIN WITH MONTE CARLO METHOD AND TEMPLATE METHOD

The modern neuroelectrical physiology indicates that there are certain patterns inneural spike train which appear more frequently than others.These favored patterns(FP)maybe related to the mechanism of neural information processing in central nervous system.Inthis paper,the quantized Monte Carlo method and template method used in favored patternrecognition are described.The Monte Carlo method determines favored patterns or the possiblefavored patterns,while the template method is used for choosing the FP from the candidates andcounting the number of the FP in the spike train.This work will be pursued for extracting theFP in some complicated spike train.     

GaN基发光二极管芯片光提取效率的研究

基于蒙特卡罗方法模拟分析了限制GaN基发光二极管(LEDs)芯片光提取效率的主要因素。结果表明,GaN与蓝宝石之间的较大折射率差别严重限制了芯片光提取效率的提高,通过蓝宝石背面出光比通过p型GaN层的正面出光的芯片光提取效率至少高20％;同时,低GaN光吸收系数、高电极反射率以及环氧树脂封装可以有效的增加芯片光提取效率,并且LEDs芯片尺寸在400μm以下时光提取效率较高。     

电子束液态曝光技术中的穿透深度预测研究

穿透深度预测是电子束液态曝光技术中的一个重要课题。采用对比分析的方法对MonteCarlo模拟和Grun公式在预测液体抗蚀剂穿透深度中的适用性进行了研究,得出了可以使用这两种方法对高能电子在液体抗蚀剂中的穿透深度进行预测的结论,并以环氧618为液体抗蚀剂进行实验验证,证明了结论的正确性,拓宽了两种方法的适用范围,为不同厚度液体抗蚀剂涂层穿透所需加速电压的快速、准确选定提供了参考依据。     

基于蒙特卡罗方法的波动水面对激光水下目标探测的影响

机载海洋激光雷达在探测水下目标时,水面波动会影响向下传输的激光光束质量。基于蒙特卡罗方法定量分析了光子在目标平面上的空间分布,以及光子返回海气界面的空间分布和时间分布。模拟结果表明,随着风速的增加,光子在目标平面上的空间分布越来越偏离高斯分布;而光子返回海气界面的空间分布则严重偏离高斯分布,且风速的影响已经不再占主导地位。根据计算获得的光子返回海气界面的时间分布,发现当风速较大时,激光雷达的探测深度降低。     

对流层散射通信信道建模及系统性能仿真

对流层散射通信是一种利用对流层媒质的不均匀性来实现超视距通信的方式.在分析对流层散射信道的统计特性的基础上,根据WSSUS假设建立了快衰落特性下的散射信道仿真模型,并进行了系统性能的实验仿真.同时通过理论计算,得出了该信道误码率的理论曲线,并与仿真结果进行比较,证明了对流层散射信道模型的合理性,该模型为分析系统的性能提供了一个有效的系统平台.     

网络可靠性仿真

讨论了在链路以及节点均可能失效情况下,把节点的失效等效为与之相连链路的失效,运用蒙特卡洛法对网络可靠性进行仿真。并给出了仿真次数对仿真结果的误差的影响。     

Monte Carlo Analysis of Yield and Performance of a GaAs Flash ADC

Monte Carlo methods are used to analyze yields and performance of GaAs flash ADCs.Due to the nonuniformity of threshold voltage,the DNL and INL of flash ADC will decrease approximately linearly.And the higher the resolution of ADC is,the faster these key nonlinear parameters decrease.When the nonuniformity increases to some degree,the yields of GaAs flash ADCs will decrease exponentially,and the missing code will increase more quickly for the higher resolution ADCs.So,GaAs HBT and HEMT with technology of etching stop will be widely used in high speed and high resolution ADCs.     

随机工艺变化下互连线ABCD参数建模与仿真

集成电路的不断发展使得互连线的随机工艺变化问题已经成为影响集成电路设计与制造的重要因素。基于电报方程建立了工艺变化下互连线的分布参数随机模型,推导出互连线ABCD参数满足的随机微分方程组,并提出了基于蒙特卡洛法的互连线ABCD参数统计分析方法,通过对ABCD参数各参量系数的正态性进行偏度-峰度检验,给出了最差情况估计。实验结果表明所提出的互连线随机模型及统计分析方法可以对工艺变化下的互连线传输性能进行有效的评估。     

Bayesian separation of spectral sources under non-negativity and full additivity constraints

This paper addresses the problem of separating spectral sources which are linearly mixed with unknown proportions. The main difficulty of the problem is to ensure the full additivity (sum-to-one) of the mixing coefficients and non-negativity of sources and mixing coefficients. A Bayesian estimation approach based on Gamma priors was recently proposed to handle the non-negativity constraints in a linear mixture model. However, incorporating the full additivity constraint requires further developments. This paper studies a new hierarchical Bayesian model appropriate to the non-negativity and sum-to-one constraints associated to the sources and the mixing coefficients of linear mixtures. The estimation of the unknown parameters of this model is performed using samples obtained with an appropriate Gibbs algorithm. The performance of the proposed algorithm is evaluated through simulation results conducted on synthetic mixture data. The proposed approach is also applied to the processing of multicomponent chemical mixtures resulting from Raman spectroscopy.