PID_Controller

压缩包中是几款LDA人脸特征识别的程序，和文章，一共是4个程序吧，有MATLAB的还有C++的(Compressed packet is LDA several facial feature recognition procedures, and articles, a total of four procedures you have MATLAB are C++ the)

DES3200实验系统的模拟信号输出通道产生一个1KHz 的方波,然后利用信号输入通道对产生的方进行低通滤波,得到一个1KHz 的正弦信号，并使用CCS 的图形显示工具显示输入和输出的波形。这里我们使用的是一个46阶的对称结构的FIR低通滤波器，其采样频率Fs为25KHZ，通带截止频率 1KHZ，阻带截止频率为3KHZ，阻带衰减为-60dB。(DES3200 experimental system analog output channel generates a 1KHz square wave, and then use the signal input channels that produce low-pass filtering the square to get a 1KHz sinusoidal signal, and use the CCS' s graphical display tool shows the input and output waveforms. Here we use a symmetrical structure of 46-order FIR low-pass filter, the sampling frequency Fs to 25KHZ, passband cutoff frequency 1KHZ, stopband cutoff frequency 3KHZ, stop-band attenuation is-60dB.)

说明：  近场声全息学习程序，适合没有基础或少量基础的研究者学习或解决NAH问题。(Nearfield Acoustic Holographic Learning Programs and Examples)

附带四种SVM工具箱的分类以及回归算法源代码(Four SVM classification and regression algorithm toolbox)

Softmax 函数处理，softmax 用于 Deep Learning 后的分类器的实现与识别，此函数的参数经过优化，有较强的泛化能力和性能(Softmax function processing, softmax classifier Deep Learning for the realization and recognition, the parameters of this function is optimized, there is a strong generalization ability and performance)

matlab2009a 破解文件 最新版(matlab2009a crack new edition)

generates noise series e

水平集经典方法的matlab工具箱,可以省却大量精力。(level set by the classical method of Matlab toolbox, we can save a great deal of energy.)

在阵列信号处理中，信号源个数的估计是一个重要问题，文献和代码给出了基于随机矩阵理论的阵列信号个数的非参检测。(In array signal processing, the estimate of the number of signal source is an important issues, documentation and code give an non-parametric detection of array signal number based on random matrix theory.)

为了提高小波分析方法在混沌去噪中的自适应能力，对不同尺度下的小波信号设定调节因子，根据混沌序列 关联维的大小确定最优阈值。为了提高寻优效率，采用遗传算法全局自适应搜索最优阈值。利用该方法对Lorenz混沌时 间序列进行了去噪分析，结果表明所提方法是非常有效的。(For the purpose of improving adaptive performance of chaotic signals de-noising with wavelet transform，a floating parameter is set to regulate the threshold of wavelet signal on different scales according to the correlation dimension of chaotic time series．The genetic algorithm is helpful tO obtain global optimum thresholds and tO reduce much time wasted by the adaptive searching computation．De-noising for Chaotic time series generated by Lorenz system is simulated to corn— pare with other methods，and the results showed that the proposed method iS effective)