文献实现的最大似然（ML）准则应用程序

Matlab code for reading and manipulating MIDI files. Includes functions for creating MIDI files.

MTLAB程序源码源程序_Maltab在数学建模中的应用贝叶斯估计程序(bayesians madeling)

说明：  用六种算子(分别是gabor、拉普拉斯、priwitt、robert、sobel、wallis),对三种图象进行边缘检测比较(Six operators (Gabor, Laplacian, priwitt, Robert, Sobel and Wallis) are used to detect and compare the edge of three kinds of images)

有信道编码，调制，信道估计等，使用混沌与分形分析的例程，Relief计算分类权重，这个有中文注释，看得明白，使用拉亚普诺夫指数的公式，可以得到很精确的幅值、频率、相位估计。

Research of random signals. Fast Furier transform. Histograms and graphs made in Matlab. Lab work in Omsk state technical university.

good simulink program for ldpc

手写体数字识别的训练数据库（MNIST）。 收集了500多位实验者的共60000个样本。(THE MNIST DATABASE of handwritten digits Four files are available on this site: train-images-idx3-ubyte.gz: training set images (9912422 bytes) train-labels-idx1-ubyte.gz: training set labels (28881 bytes) t10k-images-idx3-ubyte.gz: test set images (1648877 bytes) t10k-labels-idx1-ubyte.gz: test set labels (4542 bytes))

说明：  稀疏条件下欠定盲分离，得到混合矩阵，恢复原信号(Underdetermined blind separation under sparse conditions)

实现数据的简单DS融合，里面包含已有的五种数据的，利用贝叶斯、随机森林、k近邻分类器分类，最后融合(Simple DS fusion of data, which contains five kinds of data, uses Bayesian, random forest, k-nearest classifier to classify, and finally fuses.)

详细介绍了微弱信号的检测方法 ,做信号检测的可以参考下其中的经典方法(Details of the weak signal detection methods, signal detection can be done under one of the classical method of reference)