迭代计算

In order to investigate the performance of Iterative Detection in Golden Code system and reduce the complexity of the Iterative Detection algorithm, the Gaussian Approximation and Minimum Mean Square Error methods are proposed to obtain the soft information of each bit.

为研究级联信道编码的Golden码的迭代检测性能及降低迭代检测算法的复杂度，提出采用高斯近似和最小均方误差算法来计算迭代检测过程中所需的比特软信息。

This new method not only reduces the compute, but also needs not apriority information of the targets compared with Wishart classifier.

该迭代方法与基于Wishart距离的迭代分类方法相比，不需要矩阵的求逆运算和矩阵的对数运算，降低了迭代过程的计算量，也不再需要目标的先验信息，扩展了其适用范围。

Since the B-spline is a locally supported function, the smoothing formula can be strictly limited in a neighborhood of each point. The smoothing surface defined by the smoothing formula has the same convexity as that of the original image. Secondly, the edge points are detected by either computing the local extremum points of the directional derivative or computing zero-crossing of the second-order directional derivative of the smoothing surface. Finally, an example of an edge detector is given.

针对反卷积视觉模型的病态性，增加了适当的约束，给出了基于低松弛迭代格式的快速自适应滤波算法，对于一个N×N输入图象，由于变换核的局部支撑性质以及低松弛迭代格式的引入，使得快速自适应滤波算法的计算复杂度降为象素个数N〓的线性量阶N〓，就最小二乘算法的复杂性而言，优于不动点迭代算法的O以及基于FFT的预处理共轭梯度算法的O。

For the requirement of instantaneity, an implicit real-time discretisation is used, and the equations are integrated in a fictitiously pseudo time. This approach allows the real-time step to have two-order accuracy and accelerates the speed of convergence. Based on the theory of Mr. John K. Dukowicz and the algorithm of Mr.

在非定常计算中，本文同时采用当地和虚拟时间步，即双时间推进，在每个真实时间步上采用虚拟时间进行定常迭代，从而不但使时间迭代具有二阶精度，并使计算时间大大缩短，这样程序与定常程序在算法思想上一脉相承。

By the principle of penalt function, the irrotational fluid field is constrained, and author has proposed the Difference Method for distinguishing fluid vortex modes among the natural modes of coupled systems. Author has calculated the real frequencies with the extra deductive method, and discussed the influence of the numerical solution due to the irrotational constraint.

对于流体有限扰动时的非线性耦合系统的有限元分析作了一些探讨，给出了非线性系统的基本方程、变分原理及有限元方程，并具体地导出了三维几何非线性流体单元的切线刚度公式，采用增量形式的Newmark预报——校正逐步积分法计算系统的动力响应，由Newton-Raphson迭代法及其修正方法求解系统的非线性方程组，还列出了不平衡力余量检查和位移迭代增量检查等迭代收敛判断准则。

Based on decomposition-coordination model of an electric power system, the general augmented Lagrange method and auxiliary problem principle are adopted to form iterative distributed transfer capability computation framework, in which the influence on the internal system exposed by external system is considered through exchanging the information of boundary nodes.

该方法通过分解－协调模型实现互联电网的解耦，采用增广拉格朗日法和辅助问题原理实现输电能力按地理分区，形成一个迭代形式的分布式输电能力计算框架，迭代中通过交换边界节点信息将互联外网信息反映到各分区的计算中，各地的子网络计算相对独立。

Theinitializations both use the cost function with nominal head. Their difference is thatcontinuous function is adopted in successive linear programming while piecewiseapproximation employs the two piecewise linear segment function.

大量计算的结果表明，初始迭代点的计算采用电厂设计水头下的出力 I特性能加快整个模型的收敛性，在逐次线性规划中采用这种方式计算的初始迭代点尤其接近问题的最优点。

Its most advantage is: it is suitable for any complex geology structure, the"iterative"ray tracing method is based on Fermat"minimum travel time principle", it uses Taylor expansion and finds the solution segmentally and iteratively, then modifies the initial ray path.

迭代射线追踪方法是根据Fermat最小旅行时原理，应用Taylor展开，分段迭代求解，再逐次修改初始射线路径，完成全射线路径的迭代求解过程，其最大优点是计算的速度快，追踪射线效率高。

We analyze the advantage of the simple iterative method over Newton method at last. The numerical examples show that the more computational time has saved by the simple iterative method than the Newton method.

最后分析了简单迭代和牛顿迭代的优劣对比，数值算例表明在同样精度下简单迭代更节约计算时间。

An iterative approach to solve high order scattering solution of a vector radiative transfer equation is newly developed.

该方法将一阶散射与高阶散射迭代结合起来，计算公式简明，可计算高阶迭代解，计算时间少。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。