kolmo

模式聚类常用算法，包括k均值，模糊算法，PCA算法等。(Clustering Toolbox，including k-means clustering dcEMGMM - fits a Gaussian mixture model to data dcFuzzy fuzzy c-means clustering dcAggagglomerative (hierarchical) clustering DoClustering call the above routines)

VRP/TSP模型，可以参考最小路径相关问题分析(VRP/TSP model, you can refer to the minimum path related problems analysis)

这个zip文件中包含的原则和双方的扩展卡尔曼滤波（EKF）和全球定位系统（GPS）算法简要说明。它的主要目的是提供一个相对容易实现的扩展卡尔曼滤波，在我们使用的输入函数的形式，而不是禁止函数处理符号表达式。它也可作为一个简要介绍了卡尔曼滤波算法GPS和学习它的意思。在为EKF的例子，我们提供的原始资料，同时使用的定位扩展卡尔曼滤波和最小二乘法的任务和解决方案。 有四个M文件和两个数据文件。 Extended_KF.m包含EKF的函数体。另一些功能和GPS的示例数据文件。(This zip file contains a brief illustration of principles and algorithms of both the Extended Kalman Filtering (EKF) and the Global Position System (GPS). It is designed primarily to provide a relatively easy-to-implement EKF, in which we use symbol expressions as the form of input functions rather than the forbidding function handles. It also serves as a brief introduction to the Kalman Filtering algorithms for GPS and a mean of learning it. In the example for the EKF, we provide the raw data and solution for the positioning task using both EKF and the Least Square method. There are four m files and two data files. Extended_KF.m contains the function body of the EKF. Others are functions and data files for the GPS example.)

A program for data modeling by SOM neural network (with GUI).

Use this GUI for fitting 2-port S-parameters using the functions from RF Toolbox

BPSK QPSK MSK 16QAM四种调制方式误码率对比(BPSK QPSK MSK 16QAM modulation error rate comparison of four kinds)

主要是基于mtlab的程序，DoIqpNo参数是机器学习的例程，预报误差法参数辨识-松弛的思想，具有丰富的参数选项,Uwcywde条件基于matlab GUI界面设计,采用了小波去噪的思想。( Mainly based on the mtlab procedures, DoIqpNo parameter Machine learning routines, Prediction Error Method for Parameter Identification- the idea of relaxation, It has a wealth of parameter options, Uwcywde condition Based on matlab GUI interface design, Using wavelet denoising thought.)

Estimador de orden reducido Control en el espacio de estados

surface Acoustic wave device for Apodized filter

k means techniques de classification non supervisée (clustering) les plus utilisées. Etant donné un entier K, K-means partitionne les données en K groupes, ou "clusters", ou "classes" ne se chevauchant pas. Ce résultat est obtenu en positionnant K "prototypes", ou "centroï des" dans les régions de l espace les plus peuplées. Chaque observation est alors affectée au prototype le plus proche (règle dite "de la Distance Minimale"). Chaque classe contient donc les observations qui sont plus proches d un certain prototype que de tout autre prototype