求解度约束最小生成树的快速近似算法

针对带有度约束的最小生成树问题,给出了一种快速近似算法.首先给出了快速近似算法的核心思想:在不违反度约束和不形成圈的前提下,每次加入权最小的边.其次给出了实现快速近似算法的具体步骤,并且证明了该算法的计算时间复杂度是图的顶点数的多项式函数,证明了算法的有效性定理.大量的数值试验表明该近似算法性能良好.最后在此算法的基础上,给出了求解TSP问题的一种快速近似算法.     

The Fitzpatrick-Neville-Type Algorithm for Multivariate Vector-Valued Osculatory Rational Interpolation

In this paper,the authors first apply the Fitzpatrick algorithm to multivariate vectorvalued osculatory rational interpolation.Then based on the Fitzpatrick algorithm and the properties of an Hermite interpolation basis,the authors present a Fitzpatrick-Neville-type algorithm for multivariate vector-valued osculatory rational interpolation.It may be used to compute the values of multivariate vector-valued osculatory rational interpolants at some points directly without computing the interpolation function explicitly.     

样条权函数神经网络的一种新型算法

针对前馈神经网络在数值插值领域的应用场合,提出了一种新型结构的神经网络及其训练算法。网络拓扑结构简单,网络训练所需的神经元个数与样本个数无关,可以简单地表示成输入、输出样本向量维数之积。算法只需训练1层权函数。训练后的权函数由三次样条函数构成,而不是传统方法(反向误差传播算法“BP”或径向基函数算法“RBF”)的常数。通过求解两组线性方程组,就可以确定具体三次样条权函数形式。不存在传统梯度下降类算法的局部极小、收敛速度慢、初值敏感性等问题。仿真实验说明此算法比传统算法(如BP、RBF)精度高、速度快。     

瑞利噪声环境中水下目标辐射信号增强算法

提出了基于高阶累积量广义函数迭代的自适应滤波算法，证明了该算法的收敛性，给出了广义函数的几种具体表示式。用该算法对瑞利噪声环境中水下目标辐射信号谱增强进行了仿真研究。结果表明：该算法具有很强的抑制瑞利(白或色)噪声的能力。因此，该算法在提高水下探测系统对水下目标的检测能力方面有重要的应用价值。     

Graph theory methods for decomposition w.r.t. outputs of Boolean control networks

This paper focuses graph theory method for the problem of decomposition w.r.t. outputs for Boolean control networks (BCNs). First, by resorting to the semi-tensor product of matrices and the matrix expression of BCNs, the definition of decomposition w.r.t. outputs is introduced. Second, by referring to the graphical structure of BCNs, a necessary and sufficient condition for the decomposition w.r.t. outputs is obtained based on graph theory method. Third, an effective algorithm to realize the maximum decomposition w.r.t. outputs is proposed. Finally, some examples are addressed to validate the theoretical results.     

An augmented BV setting for feedback switching control

<正> This paper considers dynamical systems under feedback with control actions limited toswitching.The authors wish to understand the closed-loop systems as approximating multi-scale problemsin which the implementation of switching merely acts on a fast scale.Such hybrid dynamicalsystems are extensively studied in the literature,but not much so far for feedback with partial stateobservation.This becomes in particular relevant when the dynamical systems are governed by partialdifferential equations.The authors introduce an augmented BV setting which permits recognition ofcertain fast scale effects and give a corresponding well-posedness result for observations with such minimalregularity.As an application for this setting,the authors show existence of solutions for systemsof semilinear hyperbolic equations under such feedback with pointwise observations.     

基于基2-FFT的伪码快速捕获实现新算法

讨论了基于基2-FFT的伪码快速捕获方法,论述了其中常见的数据内插处理算法并提出Sinc数据内插新方法。Sinc数据内插算法采用Sinc内插滤波器来实现数据的精确内插,以满足基2-FFT的数据点数需要。仿真表明,采用Sinc数据内插方法实现的基2-FFT伪码快速捕获系统,其捕获性能优于传统的补零法和线性内插法。Sinc内插算法更适合在低信噪比下工作和对捕获时间有严格要求的系统。     

A Compromise Approach to Lexicographic Optimal Solution in Multiple Objective Programming

ＡＣｏｍｐｒｏｍｉｓｅＡｐｐｒｏａｃｈｔｏＬｅｘｉｃｏｇｒａｐｈｉｃＯｐｔｉｍａｌＳｏｌｕｔｉｏｎｉｎＭｕｌｔｉｐｌｅＯｂｊｅｃｔｉｖｅＰｒｏｇｒａｍｍｉｎｇＸＵＪｉｕｐｉｎｇＤｅｐａｒｔｍｅｎｔｏｆＡｐｐｌｉｅｄＭａｔｈｅｍａｔｉｃｓ，Ｃｈｅｎｇｄ...     

An algorithm for the discretization of an ideal projector

Ideal interpolation is a generalization of the univariate Hermite interpolation. It is well known that every univariate Hermite interpolant is a pointwise limit of some Lagrange interpolants. However, a counterexample provided by Shekhtman Boris shows that, for more than two variables, there exist ideal interpolants that are not the limit of any Lagrange interpolants. So it is natural to consider: Given an ideal interpolant, how to find a sequence of Lagrange interpolants (if any) that converge to it. The authors call this problem the discretization for ideal interpolation. This paper presents an algorithm to solve the discretization problem. If the algorithm returns “True”, the authors get a set of pairwise distinct points such that the corresponding Lagrange interpolants converge to the given ideal interpolant.     

An Improved Early Termination Sparse Interpolation Algorithm for Multivariate Polynomials

This paper presents an improved early termination algorithm for sparse black box multivariate polynomials, which reduces the interpolation problem into several sub-interpolation problems with less variables and fewer terms. Actually, all interpolations are eventually reduced to the interpolation of a list of polynomials with less terms than that of the original polynomial. Extensive experiments show that the new algorithm is much faster than the original algorithm.     

利用高阶累积量的盲多输入多输出系统辨识

针对多输入多输出线性系统(FIR MIMO)的盲辨识问题,提出了一个线性的基于方程误差的高阶累积量(HOS)算法的改进算法。该算法利用一组输出信号的四阶累积量矩阵的零空间,把一个未知多输入多输出(MIMO)信道的冲激响应辨识成一个常的单项矩阵。对于信道长度一致的不同用户的MIMO系统来说,算法只需要很弱的辨识条件。和原算法相比,改进算法充分利用了输出信号的累积量矩阵固有结构,从而提高了算法的估计性能。计算机仿真验证了算法的有效性。     

Resultant elimination via implicit equation interpolation

It is well known that resultant elimination is an effective method of solving multivariate polynomial equations. In this paper, instead of computing the target resultants via variable by variable elimination, the authors combine multivariate implicit equation interpolation and multivariate resultant elimination to compute the reduced resultants, in which the technique of multivariate implicit equation interpolation is achieved by some high probability algorithms on multivariate polynomial interpolation and univariate rational function interpolation. As an application of resultant elimination, the authors illustrate the proposed algorithm on three well-known unsolved combinatorial geometric optimization problems. The experiments show that the proposed approach of resultant elimination is more efficient than some existing resultant elimination methods on these difficult problems.     

基于对应点的三维医学图像相关性插值

现有的插值方法在进行医学断层图像插值时，要么不能兼顾灰度和形状的变化，要么计算量太大。为解决这一问题，文中提出一种基于对应点的三维医学图像相关性插值算法。通过对两幅断层图像进行门限分割，获得体素的分割值。在相同密度物质的区域内，采用体素的相关性来进行插值，不同密度物质区域采用缩放区域大小作为插值数据，使新的图像不仅在灰度上，而且在组织形状上，介于原来的断层图像之间，满足了医学图像插值的要求。与线性插值相比，新算法的视觉效果好，计算误差小；与小波插值相比，新算法的计算量极大地减少。插值结果可有效地应用于构建三维体模型.     

Distributed Algorithm for Robust Resource Allocation with Polyhedral Uncertain Allocation Parameters

This paper studies a distributed robust resource allocation problem with nonsmooth objective functions under polyhedral uncertain allocation parameters. In the considered distributed robust resource allocation problem, the (nonsmooth) objective function is a sum of local convex objective functions assigned to agents in a multi-agent network. Each agent has a private feasible set and decides a local variable, and all the local variables are coupled with a global affine inequality constraint, which is subject to polyhedral uncertain parameters. With the duality theory of convex optimization, the authors derive a robust counterpart of the robust resource allocation problem. Based on the robust counterpart, the authors propose a novel distributed continuous-time algorithm, in which each agent only knows its local objective function, local uncertainty parameter, local constraint set, and its neighbors’ information. Using the stability theory of differential inclusions, the authors show that the algorithm is able to find the optimal solution under some mild conditions. Finally, the authors give an example to illustrate the efficacy of the proposed algorithm.     

非正态数据下商业银行信用风险和经济资本度量

信用风险和经济资本度量是商业银行风险管理最重要的目标之一. 通过使用Johnson变换解决非正态数据情况下经济资本的计算问题, 克服以往研究中在Copula方法下进行Monte Carlo模拟时对正态或t分布要求的局限性, 将实际数据转换为标准正态分布, 非常方便地使用Monte Carlo模拟度量违约时间和计算经济资本. 研究结果表明, 基于Johnson变换下的Copula方法可行而且合理, 该研究为我国商业银行在《巴塞尔新资本协议》(Basel II)下进行有效的风险管理提供一些参考和新的思路.     

启发式交叉求解TSP问题的混合遗传算法

在给出度约束最小生成树的快速生成方法的基础上，设计了一种启发式交叉求解TSP问题的混合遗传算法．该算法在交叉操作的设计上，与其他遗传算法有本质的不同，该交叉操作是在不违反度约束和不形成圈的前提下，每次从父代基因所拥有的边中加入权最小的边，从而形成子代．利用该算法得到了TSP CHN144问题迄今为止最好的解．     

Normalization in Riemann Tensor Polynomial Ring

It is one of the oldest research topics in computer algebra to determine the equivalence of Riemann tensor indexed polynomials. However, it remains to be a challenging problem since Grbner basis theory is not yet powerful enough to deal with ideals that cannot be finitely generated. This paper solves the problem by extending Grbner basis theory. First, the polynomials are described via an infinitely generated free commutative monoid ring. The authors then provide a decomposed form of the Grbner basis of the defining syzygy set in each restricted ring. The canonical form proves to be the normal form with respect to the Grbner basis in the fundamental restricted ring, which allows one to determine the equivalence of polynomials. Finally, in order to simplify the computation of canonical form, the authors find the minimal restricted ring.     

基于虚拟阵元内插的互质阵列目标DOD和DOA联合估计算法

传统算法通常采取舍弃互质阵列的“差联合”阵列形成离散虚拟阵元,只利用其中连续虚拟阵元进行离波方向角(direction of departure, DOD)和波达方向角(direction of arrival, DOA)联合估计,存在自由度提升受限、估计性能不佳等问题。对此,提出基于虚拟阵元内插的互质阵列目标DOD和DOA联合估计算法。首先,将两个互质子阵以零点为中心布列,分别构成双基地多输入多输出(multiple input multiple output, MIMO)雷达的发射阵列和接收阵列,该布阵结构将传统的虚拟阵元由阵列“差联合”结构形式变成“和联合”结构形式,降低了虚拟阵列的冗余度。其次,在形成的虚拟阵元基础上,通过在虚拟阵列孔洞位置内插虚拟阵元使其连续,对于内插的虚拟阵元无实际接收信号问题,基于最小化核范数优化理论,采用协方差矩阵Toeplitz化重建的方式恢复内插虚拟阵元的等价接收信号,利于所有虚拟阵元层面的角度联合估计。最后,针对因角度配对导致的高运算量问题,结合降维多重信号分类(reduced dimension multiple signal classification, RD-MUSIC)算法使角度自动配对,从而减小算法运算复杂度。有效提高了目标分辨力和角度联合估计性能,仿真实验验证了算法的有效性。     

基于插值-拟合-迁移学习算法的机载设备故障概率预测

针对不同工作环境的机载设备故障概率预测问题,提出自适应权重的插值-拟合-迁移学习 (interpolation-fitting-transfer learning, ITF)算法。算法根据数据量和数据特征(分布相似度和信息熵)对插值、拟合、迁移学习赋予一定的权重进行线性组合。插值和拟合方法可以对故障频率进行平滑,而迁移学习可以规避数据贫化所引起的预测风险。分析该方法的可行性,通过仿真实例展示算法在预测准确度上的优势,并讨论算法中仍待解决的问题和下一步的工作。     

MIMO雷达反向投影成像算法

针对多输入多输出(multiple-input multiple-output, MIMO)雷达的多收发固定阵列结构,在标准反向投影(back projection, BP)成像算法基础上,提出了一种修正BP成像算法。该算法首先对MIMO雷达距离压缩回波数据进行时延曲线校正处理,而后沿方位向直接相干叠加各路回波处理数据,从而无需距离插值即可实现方位聚焦。与标准BP算法相比,修正BP算法大大降低了运算量,同时也克服了现有快速BP算法大都没有考虑距离插值运算的缺陷。MIMO雷达外场实测数据处理表明：与标准BP算法相比,修正BP算法有效节省了运算时间,并保持了雷达成像质量。