AugmentedLagrangianAlgorithms

盲源分离算法，该算法能够从混合信号中分离出原始信号。已用MATLAB验证：程序从正弦信号和随机噪声信号的混合信号中成功地分离出了正弦信号和噪声信号。程序保证能正确运行。(Blind source separation algorithm, which can be separated from the mixed signal the original signal. Has been verified with MATLAB: Program from the sinusoidal signal and random noise signal, mixed-signal successfully isolated sinusoidal signal and noise signal. Program guaranteed to run correctly.)

简单的自己写的matlab例程，适合初学者使用下载快速掌握图像处理。(Matlab write their own simple routines for beginners download quickly master image processing.)

基于简单一元函数的遗传算法，通过使用Shdffield工具箱，更好的实现了遗传算法的每一步，并将其结果以图的形式展现出来。代码清晰易懂。(Functions of one variable based on a simple genetic algorithm, by using Shdffield toolbox to achieve better every step of the genetic algorithm, and the results show up in the form of graphs. Code is clear and understandable.)

matlab程序求图像的梯度，获得图像的纹理信息丰富的点或边缘点、角点等。(to get the gradient of mage, the image texture information rich points or edge, corner, etc.)

RBF方法对一维函数进行插值实例，选用的函数为MQ函数，可与高阶多项式插值对比，说明RBF方法克服了龙格振荡问题。(RBF method for one dimentional function interpolation, where the Runge oscillation is disappear.)

多目标跟踪，PHD滤波器的使用，实现过程采用高斯混合方法(Multi-target tracking, using PHD filter realization method using Gaussian mixture)

对具有随机噪声的二阶系统的模型辨识，进行标幺化以后系统的参考模型差分方程为 (7.90) 式中， 为随机噪声。由于神经网络的输出最大为1，所以，被辨识的系统应先标幺化，这里标幺化系数为5。利用图7.5正向建模(并联辨识)结构，神经网络选用3-9-9-1型，即输入层i，隐层j包括2级，输出层k的节点个数分别为3、9、9、1个； (Model identification of two order systems with stochastic noise)

本程序是用matlab写的，实现对图像进行二维预测与Lloyd-Max量化压缩。是有损压缩。附有程序所用的图片。(The program is written in matlab to achieve two-dimensional projections of images and Lloyd-Max Quantization. Is a lossy compression. With a program with pictures.)

此源代码仍然是训练RBF网络的，但用的算法是梯度下降法，算法仍然是自己写的(This source code is still training RBF network, but the algorithm is gradient descent algorithm, the algorithm is still wrote it myself)

signal detection in 4x4 system using qrdmdfe algorithm with t=1,also includes lqrdm dfe algorithm