ImageCompression

最小二乘支持向量机分类预测,基于matlab的编程，实现预测分类(Least squares support vector machine classifier prediction matlab-based programming, forecasting classification)

stats 用于检验回归模型的统计量，有三个数值：相关系数r2、F值、与F对应的概率p.相关系数r2越接近1，说明回归方程越显著；F > F1-α（k，n-k-1）时拒绝H0，F越大，说明回归方程越显著；与F对应的概率p 时拒绝H0，回归模型成立.(stats for the test statistic of the regression model, there are three values: the correlation coefficient r2, F value, and F corresponds to the probability p. The correlation coefficient r2 more close to 1, note the more significant regression equation F> F1-α (k, nk-1) refused to H0, F the bigger shows the more significant regression equation F corresponds with the probability p refused H0, set up a regression model.)

主要介绍了svm的几种算法的中的具体不同和差别,随让内容简短,但对初学者来说是非常宝贵的(Mainly introduce the SVM algorithm in several different and specific differences, with the brief for content, but for beginners it is very valuable)

programa que genera figuras aleatorias

说明：  咬尾卷积码的MATLAB代码，讲的非常详细和深入(Biting tail convolutional code matlab code, said very detailed and in-depth)

在matlab中计算单位冲击脉冲响应曲线，冲击脉冲有5种，初始条件为0(in Matlab calculations unit impulse response curve, the impact of a pulse 5, the initial conditions 0)

The local binary pattern (LBP) is a simple yet very efficient operator which labels the pixels of an image by thresholding the neighborhood of each pixel and considers the result as a binary number. Perhaps the most important property of the LBP operator in real-world applications is its invariance against monotonic gray level changes caused, for example, by illumination variations. Another equally important is its computational simplicity, which makes it possible to analyze images in challenging real-time settings. LBP is also very flexible: it can be easily adapted to different types of problems and used together with other image descriptors

An ad-hoc network with a sender, a destination and a third station act- ing as a relay is analysed. The channels are modelled containing thermal noise, Rayleigh fading and path loss. Di® erent combining methods and di- versity protocols are compared. The amplify and forward protocol shows a better performance than the decode and forward protocol, unless an error correcting code is simulated. To combine the incoming signals the chan- nel quality should be estimated as well as possible. Information about the average quality shows nice bene¯ ts, and a rough approximation about the variation of the channel quality increases the performance even more. What- ever combination of diversity protocol and combining method is used second level diversity is observed. The relative distances between the relay and the stations has a large e® ect on the performance.

Threshold choice of non-coherent detection of on=-off keying (OOK) signal

Quinlan的决策树程序C4.5（M语言编写）(the program for decision tree C4.5 purposed by Quinlan )