dosc

lingjian.m-----蒙特卡罗方法 lingjian.m使用零件初始值，用蒙特卡罗方法算出总费用。其中使用了自己编制的正态分布随机数发生器产生正态分布随机数。lingjian.m是对蒙特卡罗方法的一次练习。(Monte Carlo method lingjian.m----- using parts lingjian.m initial value, using the Monte Carlo method to calculate the total cost. Using its own random number produced by the normal distribution normal distribution random number generator. lingjian.m Monte Carlo method is a practice.)

lab3 : Spatial Filtering

实验一 双极性矩形随机信号的归一化功率谱密度一 1.1 功率谱密度简介 平稳过程的任何一个非零样本函数的持续时间为无限长，显然都不满足绝对可积和总能量有限的条件。因此，它的傅里叶变换不存在即没有频谱函数。所以我们用功率谱密度来表述其频谱特性。 随机过程的任一实现是一个确定的功率型信号。而对于任意的确定功率信号f(t)，它的功率谱密度为： 式中， 是f(t)的截短函数 对应的频谱函数。f(t)是平稳随机过程 的一个实现。而随机过程某一个实现的功率谱密度不能作为过程的功率谱密度。过程的功率谱密度应该看作是任一实现的功率谱密度的统计平均，即 虽然该式给出了平稳随机过程的功率谱密度，但我们通常都不利用这个式子来计算功率谱。我们知道，确知的非周期功率信号的自相关函数与功率谱密度是一对傅里叶变换。对于平稳随机过程，也有类似的关系，即 和 对于平稳随机过程我们通常先求出其自相关函数再利用上式求出其功率谱密度。 1.2 实验要求  1.了解平稳随机信号功率谱的概念及计算方法  2.仿真不同占空比，等概、非等概双极性矩形随机信号的归一化功率谱密度  3.分析不同信号所包含的频谱分量，有无直流分量和定时分量信息 (A pair of rectangular random experiment polarity signal sample return a nonzero function of any duration power spectral density of a 1.1 power spectral density of a stationary process Introduction of infinite length, apparently not satisfied absolutely integrable and the total energy of the limited conditions. Therefore, it is the Fourier transform of the spectrum that does not function does not exist. Therefore, we use the power spectral density to express their spectral characteristics. Any random process is a realization of a certain type signal power. For arbitrarily determined power signal f (t), its power spectral density is: wherein, is f (t) is a function corresponding to the truncated spectrum function. f (t) is a stationary random process of realization. The random process to achieve a certain power spectral density can not serve as the power spectral density of the process. Power spectral density process should be seen as a statistical average power spectral density of any r)

程序包含的设置包括：列数、样本总个数、建模样本数、预报因子数、预报对象数、学习因子、 动量因子、训练次数、总体误差、隐节点数。(Contains the settings include: the number of rows, the total number of samples, sample size modeling, forecasting factor number, forecasting the number of objects, learning factor, momentum factor, training frequency, the overall error, hidden nodes.)

应用matlab生成白噪声和有色噪声的程序设计(failed to translate)

龙格库塔法matlab仿真代码。适于学习控制的朋友在进行matlab仿真时学习方法本质。(Runge-Kutta method matlab simulation code. Learning Control for friends during the matlab simulation method when the nature of learning.)

ofdma的一个简单的收发程序采用QPSK调制噪声是高斯白噪声(A simple transceiver of ofdma program using QPSK modulation noise is gaussian white noise)

计算 两个Guassian Mixture Model的KL-Divergence(calculating the KL-divergence between two Gaussian Mixtures)

利用遗传算法优化PID参数，自动搜索最有 的PID参数(Using genetic algorithm to optimize PID parameters and automatically search the most PID parameters)

说明：  模式识别PCA(principle component analysis)源码.matlab 格式。PCA为经典而且经常使用的算法。(pattern recognition PCA (principle component analysis) source. Matlab format. PCA to the classic and often use the algorithm.)