VQmatlab

详细阐述了设计通信网络仿真环境里的各个参数，如带宽，延迟，掉包率，等等因素，这里的网络包括了WLAN，3G以及各种业务所需的基本条件(Detail the design of communication network simulation environment, all parameters such as bandwidth, delay, substitution rate, among other factors, where the network includes a WLAN, 3G, and the basic conditions required for various services)

this folder contai luekemia data set

Compressive Sampling是最近讨论非常热烈的数据处理方法。用一个与变换基不相关的观测矩阵将变换所得到的高维信号投影到一个低维空间上，然后通过优化求解从这些少量的投影中以高概率重构出原信号。(Compressive Sampling is a very lively discussion of recent data processing method. A base with the transformation matrix will not change the relevant observation obtained a high-dimensional signal projected onto low-dimensional space, and then solved by optimizing the projection from these small reconstructed with high probability in the original signal.)

对盲源分离中独立成分分析法的实现，包括去均值，白化，和牛顿迭代法进行的ICA分析。(The implementation of independent component analysis in blind source separation includes ICA analysis of mean value, whitening and Newton iteration.)

支持向量机导论 matlab中支持向量机算法的入门级经典教材(Introduction to matlab support vector machine support vector machine algorithm is the classic entry-level teaching materials)

基于自回归,全极点的AR模型,对随机信号的进行的功率谱估计.(based on the return of all Poles AR model of random signal of the power spectrum estimation.)

Implementation algorithm to change contrast of an image

TSP matlab遗传算法求解，关于商旅问题(TSP matlab genetic algorithm)

说明：   改进的牛顿法求解： 方程的构造方法:给出[0,1]区间上的随机数(服从均匀分布)作为方程的根p*. 设你的班级数为a3,学号的后两位数分别为a2与a1,从而得到你的三次方程 例如:你的31班的12号,则你的方程是21x3+60x2+2x+a0=0的形式. 方程中的系数a0由你得到的根p*来确定. (improve Newton's method : Construction of the equation : given interval [0,1] on the random number (subject to uniform distribution) as the root equation p*. set up your classes at a3, after learning of the double-digit for a1 and a2, so you get the three equations for example : Your 31 classes on the 12th, then your equation is 21x3 60x2 2x a0 = 0 forms. the equation coefficients a0 you get from the root of p* to determine.)

优化的基线学习机，收敛速度快，可用于分类和拟合(The baseline learning machine optimization, fast convergence speed, and can be used for classification and curve fitting)