sig_shift

噪声环境下的予以增强程序，用小波函数 db6 对信号进行3层分解(Noise environment to be enhanced procedures DB6 wavelet function with the signal 3-layer decomposition)

这是一个自适应的遗传算法实例， 是在遗传算法的基础上改进的， 例子是和matlab工具箱与应用书上的相似(This is an example of adaptive genetic algorithm is based on the genetic algorithm improved, for example, is matlab toolbox and application of similar books)

gmsk仿真程序，相干解调实现方法。。还是很好用的。。。(gmsk simulation program, to achieve coherent demodulation method. .)

This package contains two implementations of the Generalized Cross Correlation based on the Phase Transform, the first one is applied to TDOA and the second one is more general where the GCC is averaged over n blocks.

共轭梯度法最优化算法matlab源程序。。(Conjugate Gradient Method optimization algorithm matlab source code. .)

用MATLAB编写的读取视频图片的程序。(Using MATLAB program for reading video picture.)

无刷电机电磁设计程序,包含空载和负载时的磁路工作点的校正(bldc design )

本程序实现对图像的增强，通过模糊增强算法，使图像的边缘信息更加明显，图像出来，能够更好的处理高频信息(This procedure to achieve an image enhancement, through the fuzzy enhancement algorithm, so that the edge of the image information more clearly, the image out, better able to deal with high-frequency information)

Genetic algorithm feature selection by using Logistic Regression

说明：  本文提出了一种基于混沌加密的DCT域二值图像数字水印算法.该算法首先利用混沌序列将含有版权信息的有意义二值图像水印进行加密,然后利用人眼视觉系统(HVS)的特性将原始载体图像进行块分类,最后将不同强度的水印分量嵌入到不同类的图像块DCT中频系数中去.利用混沌序列对水印进行加密,使水印本身具有安全性,对数字作品进行双重保障.通过高斯噪声,椒盐噪声以及JPEG压缩和剪切等攻击试验,表明了该算法具有较好的鲁棒性.(Encryption in DCT domain is proposed. Firstly, the meaningful binary image watermark containing copyright information is encrypted by using chaotic sequence, and then the original image is classified by using the characteristics of human visual system (HVS), Finally, the watermark components with different intensities are embedded into the DCT intermediate frequency coefficients of different types of image blocks. Chaotic sequences are used to encrypt the watermark, so that the watermark itself has security and double protection for digital works. Experiments on Gaussian noise, salt and pepper noise, JPEG compression and clipping show that the algorithm has good robustness)