国外分享的基于OFDM的认知无线电系统资源分配测试源程序

遗传算法与神经网络结合在股票预测中得到很好的应用 (Genetic algorithms and neural network prediction in the stock get a good application)

小时宽带宽积信号的脉压算法（lfm和gmsk信号）本文对项目使用的LFM、Taylor编码两种不同小时宽带宽积信号的实际回波波形和特性进行了分析，并针对性地使用了倒推法和循环迭代法，获得优化的脉冲压缩系数，提高了脉冲压缩的主副瓣比。试(Applications of two digital pulse compression arithmetics for small TB signal)

bp网络的一个例子，考虑两单元输入的联想问题(bp example of a network, consider two modules imported Lenovo problems)

database management system jdbc related

EMD端点效应参考代码，包含多个算法：极值平均，边界特征尺度延拓，极值延拓，基于ISBM延拓，平行延拓，镜像延拓(EMD end effect reference code contains multiple algorithms: extreme average, the boundary characteristic scale extension, extreme extension, extension based ISBM parallel extension, mirror extension)

说明：  假定导引头工作在理想状态，能够无延迟无误差地输出与视线角速度成比例的信号u，在此情况下的导引律(Assuming that the seeker works in an ideal state, it can output the signal U proportional to the angular velocity of the line of sight without delay and error. In this case, the guidance law)

Harmony search algorithm . a method for optimization a nonlinear problems with constraints.

说明：  通过把分散在时间轴上的离散数据看成一组连续变化的序列，采用累加和累减的方式，将灰色系统中的未知因素弱化，强化已知因素的影响程度，最后构建一个以时间为变量的连续微分方程，通过数学方法确定方程中的参数，从而实现预测目的。(By regarding the discrete data scattered on the time axis as a series of continuous changes, and adopting the way of accumulation and subtraction, the unknown factors in the grey system are weakened, and the influence degree of the known factors is strengthened. Finally, a continuous differential equation with time as variable is constructed, and the parameters in the equation are determined by mathematical method, so as to achieve the prediction purpose.)

【实例简介】

这是一个简单的FFT为无电位1 +1 +1 Ð 薛定谔方程的光束传播方法。如果该软件灵活，允许引入的术语（如果是极少数需要包括色散效应高阶导数）。例如，如果一个人渴望解决的一个方程的形式： （四/ dz的+ Ð ^ 2/dx ^ 2 - 0.25 * Ð ^ 3/dx ^ 3）== 0 Ÿ 它可以解决使用以下代码： Ž = linspace（0,1,512） x = linspace（-5,5,1024） psi0 =进出口（- x的。^ 2） Ð = D_lateral1D（十，2）-0.25 * D_lateral1D（十，3） 字段= LinearBPM（psi0，有D，Z） 凡D_lateral1D是一个函数，提供“取代”的微分算子。 这个函数是在一个非线性的BPM，我曾对（分步，谱方法等），将提交给我，当我完成他们的文件和组织功能更复杂的设置非常初期阶段。 (This is a simple FFT based beam propagation method for potential-free 1+1+1D Schrodinger equation. The software if flexible and allows to introduce high-order derivatives in the term (very handful if is needed to include dispersion effects). For example, if one desires to solve an equation in the form: (d/dz+ d^2/dx^2- 0.25*d^3/dx^3)Y == 0 It can be solved using the following code: z=linspace(0,1,512) x=linspace(-5,5,1024) psi0=exp(-x.^2) D=D_lateral1D(x,2)-0.25*D_lateral1D(x,3) field=LinearBPM(psi0,D,z) Where D_lateral1D is a provided function that "replaces" the derivative operator. This function is a very early stage in a more complex set of non-linear BPM functions that I have made (split-step, spectral methods, etc.) and I will submit them when I finish their documentation and organization. Sorry for my English, I m eager for your comments. )