control

矢量算法，计算声源位置。速度比较快，精度很高。(Vector algorithm to calculate the location of a sound source. Relatively fast speed, high precision.)

语音端点检测算法，在matlab里面实现语音端点的准确检测。为读者开发噪声环境下的精确VAD提供思路.(voice endpoint detection algorithm in Matlab voice inside the precise endpoint detection. Readers development under noisy environments to provide accurate VAD ideas.)

基于matlab的语音回波抵消系统仿真设计程序原件及说明(Voice echo cancellation system based on matlab simulation program original and instructions )

这是用itpp库编写的卷积码的解码程序。由于用C++库和Intel的MKL库，所以速度飞快。同样的仿真点数， Matlab2012下用了15分半钟，而这个程序仿真只用了半分钟，相当于是matlab速度的30多倍。(This is itpp prepared for the convolutional code decoding process. Because with C++ libraries and Intel' s MKL library, so fast. The same simulation points under Matlab2012 spent 15 and a half minutes, and this program simulation with only half a minute, the equivalent of 30 times the speed of matlab.)

说明：  使用ID3决策树算法预测销量高低,对数据进行预处理，最终打印并画决策树(Using ID3 Decision Tree to Predict Sales)

matlab一本学习非常好的书，大家可以读一下，快速入门(matlab a learning very good book, we can read the Quick Start)

MATLAB实例，对学习有用,MATLAB实例，对学习有用MATLAB实例，对学习有用MATLAB实例，对学习有用(MATLAB is good for you,thank.MATLAB is good for you,thank.MATLAB is good for you,thank.)

说明：  四轮驱动的自动驾驶车辆前楼周轮胎力优化分配(Optimization of tire force distribution around the front floor of four-wheel drive vehicle)

将图像分为nxn的分块,之后将各分块三个颜色通道的直方图平均化,最后进行离散余弦变化,分别提取各通道低频系数(The image into nxn sub-block after block of all three color channels of the histogram equalization, and finally the discrete cosine change, were extracted from the low frequency coefficients of each channel)

本文算法是将24 位真彩图的亮度分量分块 DCT ,然后利用DCT 量化系数的奇偶特性嵌入8 位灰度水印图,并通过PN 序列加密来保证水印的 安全性。算法中,将灰度图像转换为二进制数据 流,置乱后扩频调制,隐藏在宿主DCT 系数量化值 中,提取时不需要宿主信号,实现水印信号的盲提 取。(This algorithm is 24-bit true color pictures of the luminance component of block DCT, quantization DCT coefficients and then use the parity properties of 8-bit grayscale watermark embedding map, and through the PN sequence of encryption to ensure the security of the watermark. Algorithm, the gray image into binary data stream, scrambling after the spread spectrum modulation, hidden in the host DCT coefficient quantization values, the extraction does not require the host signal, the blind watermark extraction.)