drbd-8.3.6.tar

主元分析法的C语言程序，主要用在降维的方面，也可用在信号处理，故障诊断。(Principal Component Analysis Method C language program, mainly used in dimensionality reduction, can also be used in signal processing, fault diagnosis.)

LDPC的译码算法，其中使用matlab和C的联合编程，C主要做了译码算法，并且有性能的仿真测试(LDPC decoding algorithm, including joint programming matlab and C, C major to do the decoding algorithm and performance simulation test)

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.)

matlab hann hamming window FFT DFT DTFT

针对信号自适应chirplet分解未知参数多、实现起来比较困难的特点，文献[1]提出了一种新的chirplet分解快速算法。该算法利用计算信号的二次相位函数，得到其能量分布集中于于信号的调频率曲线上的结论，此时通过谱峰检测可同时获得chirplet调频率、时间中心和幅度的估计；然后通过解线性调频技术获得其初始频率和宽度的估计，仿真结果验证了本文算法的有效性。 (For the signal adaptive the chirplet decomposition of unknown parameters to achieve difficult to literature [1] proposed a new fast algorithm for chirplet decomposition. The algorithm makes use of the secondary phase of the calculated signal function, the conclusion focused on energy distribution curve to tune the frequency of the signal by the peak detection chirplet tune frequency, time center and amplitude estimates then by solving chirp technology to obtain the initial frequency and width is estimated that the simulation results demonstrate the effectiveness of the proposed algorithm.)

4-9-1的结构ga-lm的bp算法 有示例数值 不是拟合而是预测(4-9-1 structure of ga-lm bp algorithm is not fit but there are examples of numerical prediction)

筛板精馏塔的操作与塔效率的测定，经过改良源代码(sieve plate distillation column)

说明：  VC调用matlab，实现VC与matlab混合编程，效果很好(VC call matlab, matlab mixed programming with VC to achieve good results,)

由设计要求可知，wp<=0.2π，允许误差幅度小于1dB，ws>=0.3π内衰减应大于15dB。为了验证数字滤波器的实际滤波功能，设计程序时，把输入序列定为验证函数x（t）=4sin（100πt）以及x（t）=4sin（200πt）和x（t）=4sin（400πt）。又因为Ts=1.25ms，T=80ms，故采样点为64，即N=64，所以当M=6，可以同时调试并验证程序的滤波功能。(Seen by the design requirements, wp < = 0.2π, allow a margin of error of less than 1dB, ws> = 0.3π within attenuation greater than 15dB. In order to verify the actual filtering function of the digital filter design program, the input sequence as validation function x (t) = 4sin (100πt) and x (t) = 4sin (200πt) and x (t) = 4sin (400πt) . And because Ts = 1.25ms, T = 80ms, so the sampling points is 64, that is N = 64, so when M = 6, can debug and validate filtering program.)

Tutorial Two (You should run it in the Command Window)