一种改进的基于奇异值分解的信源数目估计算法

信源数目估计问题在盲源分离中具有重要的意义。研究了传感器数目大于信源数目时的源数估计问题。首先分析了用奇异值分解法进行信源数目估计的优势与不足,然后提出了一种改进的基于奇异值分解的信源数目估计算法。该算法首先对含噪混合信号进行奇异值分解,然后检测信号分量与噪声分量之间的转折点,将信号分量与噪声分量区分开来,从而得到信号源的数目。实验仿真表明,该算法在低信噪比以及采样点数较少时仍然具有好的性能。     

星载AIS的复值FastICA算法改进

针对星载船舶自动识别系统(AIS)的含噪复值信号盲分离算法分离效果不佳的问题,提出了改进的复值快速独立分量分析算法（FastICA）。该改进算法针对混合信号数目大于源信号数目的超定情况,对含噪混合信号的协方差矩阵进行特征值分解,利用其噪声对应的几个较小特征值估计噪声方差,修正白化矩阵,再应用Huber M估计函数优化该算法的目标函数。实验结果表明,运用该算法信号均方误差(SMSE)变小,信干比(SIR)变大,提高了信号的分离性能;同时,优化后的目标函数使算法具有良好的稳健性。     

利用时频域稀疏性的LFM信号盲源分离方法

信号分离是雷达电子对抗的重要环节。考虑到雷达信号在时频域具有稀疏性的特点,在独立分量分析的基础上,提出了一种基于时频域稀疏性的线性调频雷达信号盲源分离方法。首先对混合信号进行短时傅里叶变换,在每个频点利用自然梯度算法分离信号,由分离信号幅度的比值作为对源信号后验概率的估计;然后根据相邻频点后验概率序列的相关性进行排序,确保各个频点的分离信号属于同一个源信号;最后设计时频掩码分离信号。进行了线性调频雷达信号卷积混合的盲分离实验,所提方法分离结果明显优于传统独立分量分析方法的分离结果,验证了该方法的有效性。     

一种基于频移滤波器的混合信号盲恢复算法

针对单天线接收的频谱混叠的混合信号盲恢复问题,在频移滤波器结 构上,提出了一种基于相关函数误差准则的自适应频移滤波信号盲恢复算法。该算法利用滤 波 器输出信号和参考信号以及混合信号与参考信号之间的相关函数误差来调整自适应滤波器输 出权值。分析了该算法的稳态性能。仿真结果表明：在信噪比大于-5 dB的条 件下, 该算法对混合信号分量的恢复相似度达80%以上,并且该算法仅需已知混合信号分量的循环 频率,无需其它先验信息,具有计算量小、敛速度快的优点。     

采用延时积分的MSK软判决解调方法

针对最小频移键控（MSK）信号的软信息计算,提出了一种采用延时积分的软判决解调方法。该方法建立在MSK延时判决解调方案的基础上,按照条件概率理论的推导结果,使用接收已调信号同相分量和正交分量积分度量的乘积来表征原二进制信息码元的对数似然比信息,并以此作为MSK信号的软判决输出。分析与仿真结果表明,该方法简单方便,具有较大的通用性和灵活性,在Turbo联合MSK的编码调制系统中应用可以获得较为优异的抗噪声性能。     

基于加权最小二乘支持向量机的欠定盲源分离

为了进一步提高欠定盲源分离算法中混合矩阵估计方法的性能,提出了一种基于加权最小二乘支持向量机（SVM）的欠定盲源分离混合矩阵估计方法。该方法利用信号的方向角度特征估计出有效信源信号个数,然后采用加权最小二乘支持向量机方法获得初始权值,每次将其中一个权值对应的样本点作为测试样本,其余点作为训练样本,依次对样本的误差变量进行更新,再根据权值计算公式实现所有权值的更新,进而确定最优分类平面,实现对观测信号的最优分类,最终估计出混合矩阵。实验结果表明,新算法是有效的,其平均误差是基于K-均值方法误差的0.2倍左右,是基于SVM算法平均误差的0.5倍左右。     

一种新的多信号卷积混合信号盲分离算法

对于含噪声情况下多个源信号卷积混合盲分离,由于混合矩阵比较复杂,分 离算法会出现迭代次数增加、收敛速度变慢等问题。在对多信号卷积混合进行合理简 化的基础上,提出一种以四阶累积量为独立准则的多信号卷积混合的新的时域盲源分离算法 。由于采用高阶累积量为独立准则,该算法对高斯噪声具有良好的抑制作用,改善了信噪比 。 其次,算法也建立了步长因子的选取与二次残差之间的非线性函数关系,使得算法既获得了 较 快的收敛速度,也得到较高的分离精度。仿真数据表明提出的算法对于多个源信号卷积 混合具有良好的分离效果。     

基于自适应惯性权重粒子群优化的多跳频信号盲源分离

针对跳频通信中多跳频信号的盲源分离问题,提出了一种基于自适应惯性权重粒子群的盲源分离算法。该算法将分离信号的负熵作为目标函数,依据迭代前后每个粒子适应度值间差值自适应地调节惯性权重。把适应度值变差的粒子惯性权重设成零,以消除惯性分量不利影响,这样可以减少无效迭代次数,提高收敛速度。应用于盲源分离时,比经典算法分离效果好且克服了激活函数选取难题。实验结果表明该算法用于多跳频信号盲分离时性能稳定且收敛速度快,与经典算法比较优势明显,为智能算法在盲源分离方面的研究提供了一定的参考。     

基于曲线拟合的信噪比估计校正方法

介绍了QPSK调制的DSSS接收机中基于矩估计的信噪比估计算法,以及在FPGA上实现该算法的关键技术.由于实现误差超出期望,运用了基于曲线拟合的校正方案.以估计幅度值和幅度差值作为自变量和因变量,运用曲线拟合工具获得信噪比估计值应有的修正误差.依据该修正值调整测试值,可使信噪比估计性能大幅提升.最后通过实测证明了校正算法的性能.     

采用MRSSA-ICA的共信道多通信信号盲分离

针对时频域部分重叠的多个跳频通信信号共信道盲分离问题，提出了一种新的共信道盲分离算法SCBSS（Single Channel Blind Source Separation）。首先，重新定义多分辨奇异谱分析（Multi-resolution Singular Spectrum Analysis，MRSSA）算法，利用其冗余性来重构伪多输入输出模型；接着引入独立分量分析算法用于提取感兴趣的独立分量。仿真结果验证了所提算法分离多个正交相移键控（Quadrature Phase Shift Keying，QPSK）调制的时频域部分重叠跳频通信信号的有效性和鲁棒性，且不需要任何先验。     

Double Lookbacks

A new class of options, double lookbacks , where the payoffs depend on the maximum and/or minimum prices of one or two traded assets is introduced and analyzed. This class of double lookbacks includes calls and puts with the underlying being the difference between the maximum and minimum prices of one asset over a certain period, and calls or puts with the underlying being the difference between the maximum prices of two correlated assets over a certain period. Analytical expressions of the joint probability distribution of the maximum and minimum values of two correlated geometric Brownian motions are derived and used in the valuation of double lookbacks. Numerical results are shown, and prices of double lookbacks are compared to those of standard lookbacks on a single asset.     

国内外蜡烛的发展状况

蜡烛主要由主燃剂和烛芯构成。按照主燃剂成分组成,蜡烛分为石蜡型蜡烛和非石蜡型蜡烛。蜡烛的质量取决于主燃剂,功能取决于所加入的配料。本文综述了国内外蜡烛发展状况,并对国内蜡烛工业发展上存在的诸如石蜡型蜡烛中存有气泡、蜡烛表面有鳞片和花点,非石蜡型蜡烛存在成型性和保型性不好,科技投入不够,人才匮乏等问题提出了建议。     

提高DSRC道路信标定位精度的最大信号法及其工程实现

本文概述了中国外贸出口体制的改革进展和成就,探讨了WTO对出口体制的要求及中国出口体制与之相比存在的差距,并按照WTO原则和社会主义市场经济要求,对中国入世后进一步完善出口体制提出12个方面的主要措施。     

弱暂态正弦信号的检测

将Duffing振子用于弱暂态正弦信号的检测,发现可被检测的暂态正弦信号的持续时间下限为12个周期,对相位随机分布并淹没在噪声中的弱暂态正弦信号可以用一组(16个)Duffing方程检测到,对到达时间的定位可以精确到一个周期。     

未知衰落信道中MPSK信号的非数据辅助载波频率估计方法

未知衰落信道中的MPSK信号的非数据辅助载波频率估计是通信参数估计中的复杂问 题,目前尚缺乏有效的针对性研究成果。针对该问题,提出采用先补偿信道影响,再去调制 , 然后进行非衰落信道单频信号频率估计的方法。该方法比进行频率和信道冲激响应联合估计 的方法简洁和高效,而且解决了联合估计中难以有效去调制的问题。通过和非衰落信道的频 率估计结果对比,表明其估计性能的下降小于3dB,能很好地满足应用需要。     

Participation of working women in decision‐making process as consumer

In a world where the role of women in decision‐making is seldom adequately appreciated, they make a remarkable contribution due to their hard work and sense of confidence. It is observed that women are mostly involved in repetitive and monotonous household work irrespective of the fact that they share most of family responsibilities and perform a wide range of duties in and outside home. On the other hand men perform activities, which require skills, but there is sufficient evidence, which show a clear, although slow shift of stereotype sex roles. In early societies, decision‐making was predominantly done by menfolk being the breadwinner of the family. With modernization and education women have been empowered to make the best use of human and non‐human resources in management of the family with respect to efficient use of time and energy. So, a study was undertaken to know how far the working women of Faridkot district participate in the decision‐making process with the following objectives:

• (a) to find out the level of participation of working women in the decision‐making process as consumer.
• (b) to find out the level of difficulty among working women at different steps of the decision‐making process.

The present investigation pertaining to Faridkot district of Punjab state of India was exploratory in nature. A sample of 100 working women was selected by simple random sampling procedure. The responses were recorded through a well structured and pretested questionnaire including an index developed to measure level of participation and difficulty as realized by working women in the decision‐making process. After collecting the data, it was analysed by employing simple statistical tests like frequency percentages The important findings emanated out of the present investigation are as follows:

• ? Most of the respondents were middle aged, educated up to graduation level, having nuclear family, small size of family with monthly income more than Rs.5000.00. It was also observed that majority of the respondents were in the teaching and medical professions.
• ? Working women were found to be involved more in the case of food items (79.0–93.0%) as compared to non‐food items (6.0–46.0%) with an exception of their major participation (78.0%) at the stage of identifying the problem in case of non‐food items.
• ? The study revealed that the time consumed at various steps of consumer decision‐making process is minimum for most of the steps in case of food items but moderate to maximum in case of non‐food items. But for analysing the alternatives the time consumed was moderate to maximum for both the categories.
• ? Findings also revealed that difficulty level was from moderate to high (32.0–98.0%) in most of the steps in case of food items whereas, in case of non‐food items it was from low to high (11.0–78.0%)

     

Hedging and value at risk: A semi‐parametric approach

The non‐normality of financial asset returns has important implications for hedging. In particular, in contrast with the unambiguous effect that minimum‐variance hedging has on the standard deviation, it can actually increase the negative skewness and kurtosis of hedge portfolio returns. Thus, the reduction in Value at Risk (VaR) and Conditional Value at Risk (CVaR) that minimum‐variance hedging generates can be significantly lower than the reduction in standard deviation. In this study, we provide a new, semi‐parametric method of estimating minimum‐VaR and minimum‐CVaR hedge ratios based on the Cornish‐Fisher expansion of the quantile of the hedged portfolio return distribution. Using spot and futures returns for the FTSE 100, FTSE 250, and FTSE Small Cap equity indices, the Euro/US Dollar exchange rate, and Brent crude oil, we find that the semiparametric approach is superior to the standard minimum‐variance approach, and to the nonparametric approach of Harris and Shen (2006). In particular, it provides a greater reduction in both negative skewness and excess kurtosis, and consequently generates hedge portfolios that in most cases have lower VaR and CVaR. © 2009 Wiley Periodicals, Inc. Jrl Fut Mark 30:780–794, 2010     

小子样不可修武器装备的Bayes可靠性增长分析

现有的可靠性增长模型大多是针对可修系统建立的,而现实中大多存在的是不可修产品,即产品投入可靠性增长试验出现故障后,很难修复如新再次投入试验.针对武器装备小子样、不可修的特点,提出了Bayes可靠性增长分析方法.首先利用多台产品异步增长理论得到各阶段试验数据的似然函数,然后构造了增长模型参数的Gamma-均匀验前分布并给出了模型参数估计的ML-II方法,通过Bayes统计推断对装备研制全过程的可靠性增长规律进行分析,最后通过仿真示例说明了该方法在工程上的应用.     

恶劣海况下高频信号多跳传播特性分析

保持与陆地的联系对于船只在浩瀚的海洋中航行非常重要。来自陆地的高频信号经电离层和海面多次折反射后到达船只，受到不同海况的传播衰减特性影响。针对粗糙海面反射模型，从建立海面波动方程出发，考虑到波浪屏蔽对信号的影响，建立了新的反射衰减模型。利用该模型进行仿真，得到高频信号超视距海上传播的最大跳变规律。分析了湍流海面和平静海面对高频信号衰减的差异，揭示了海浪遮挡是引起差异的主要因素。对于航行在波涛汹涌的海面上的船舶，建立了包含浪和涌特征的多路径反射叠加模型，该模型考虑了大尺度波和小尺度波的海面情况，更加接近实际情况。最后利用所建立的模型计算了高频信号的最大跳数和通信距离。     

Total Factor Productivity in the Turkish Insurance Industry

This paper employs the Malmquist total factor productivity (TFP) index to examine the total factor productivity change in the life and non‐life branches of the Turkish insurance sector for the period 2000–2005. The Malmquist productivity index is decomposed into two components: the efficiency change component and the technical change component. The results indicate a decrease of 19.4 % in total factor productivity in the life insurance sector and an increase of 6.1% in the non‐life insurance sector during the sample period. Significant TFP growth in the non‐life insurance sector has occurred during 2003–2005. This growth is seen to be mainly due to the change in production technology. The firms are then split into two groups, domestic and foreign firms, according to their ownership composition. The results indicate a decline in total factor productivity for both groups in the life insurance sector whereas in the non‐life insurance sector, an increase in the total factor productivity of domestic firms is observed.