EMD

matlab hello world程序、windows对话框的mex程序，对于学习matlab混合编程很有帮助(matlab hello world programs, windows dialog mex program, very helpful for learning matlab Hybrid Programming)

使用最大期望算法可以在概率模型中寻找参数最大似然估计。(Using the maximum expected algorithm can seek parameters of maximum likelihood estimation in the probabilistic model)

遗传算法工具箱包含安装方法，一般matlab中需要独立安装(Genetic Algorithm Toolbox)

多发多收的稀疏信道估计的仿真程序，运用LS、OMP与时反算法( Multiple overcharged sparse channel estimation simulation program, using LS, OMP and time-reverse algorithm)

这是我在网上找的支持向量机的代码，应该比较普通(support vector machine )

均匀线阵DOA估计Unitary -ESPRIT算法MATLAB程序(Unitary -ESPRIT for ULA)

说明：  LBP算法实现图像的纹理分类K (LBP algorithm of image texture classification)

这是一个半导体激光器弛豫振荡现象的仿真程序，计算参数和速率方程模型都取自经典教材Semiconductor Lasers(Agrawal 著)，可以作为学习和进一步研究半导体激光器各种动态特性的基础材料。(This is a simulation program by Matlab for calculating the oscilating in semiconductor lasers, in which the rate equation model and the parameters used are all from the classic technical monograph by Agrawal. This could be a basic material for further investigating and studying different dynamic activities of semiconductor lasers. )

ContourletSD: MATLAB code implementing the new contourlet transform with sharp frequency localization as described in the following paper. Y. Lu and M. N. Do A new contourlet transform with sharp frequency localization. Proc. IEEE International Conference on Image Processing, Atlanta, 2006. Version: October 2009

（1） 本系统用到四种神经网络对漏电嫌疑系数进行预测，分别是BP神经网络、RBF神经网络、GRNN神经网络、Elman神经网络，通过对神经网络结构参数的设置和调试，可以得到很好的预测效果，实际值和预测值之间的误差能达到5 以内。 （2） 本系统使用matlab做GUI界面，在界面上就可以分别对这四种神经网络的结构参数进行设置，从而可以调整神经网络的结构，达到较好的预测精度。 ((1) The system used four neural network to predict the suspected leakage coefficient, respectively, the BP neural network, RBF neural network, GRNN neural network, Elman neural network, the neural network structure parameter setting and commissioning, can get very good prediction, between the actual and predicted values ​ ​ of error of less than 5 . (2) the use matlab to do the GUI interface, the interface can each of these four neural network structure parameters set, the structure of the neural network, which can be adjusted to achieve better prediction accuracy.)