peizhun1

语音信号断点检测MATLAB系统仿真程序详述(Breakpoint detection of speech signals detailed MATLAB simulation program )

code rich machin chose

说明：  TSP simulated annealing matlab code

这个是用来计算白噪声测试的，特别适用于振动台试验。编程简单，测试也比较符合预测值。(This is used to calculate white noise, especially for shaking table test. The programming is simple and the test is more consistent with the predicted value.)

EIGENFREQ finds an approximation for the eigenfrequency of the resonator within a range and at a resolution specified by the user. The user specifies the range in terms of the minimum and maximum values of each of the real part and the imaginary part of frequency, and the user specifies the resolution in terms of the number of evenly-spaced real frequencies and complex frequencies at which EIGENFREQ is to performcalculations.

基于matlab的地震傅式谱编写程序，很有用的一个程序。(Fuh-based seismic spectrum matlab programming, a very useful program.)

function deconvolution in matlab

说明：  这是仿真powell算法的简单程序，已经调试过了，效果不错，请大家多提宝贵意见(This is a simple simulation algorithm powell procedures, have been debug, good results, please mention the valuable advice)

鲁棒性非常好的ICA算法程序，做独立成分分析时候效果好(robust-ICA)

when SNR is high , 1st sample estimator provides good estimate of A. it has no noise effect . so we don’t need averaging effect to reduce noise when SNR is high. However variance is still low in sample mean estimator which shows that in reality ,sample mean is good estimator than 1st sample estimator. So, for high SNR, we do not need to reduce the effect of noise by averaging. or we can say that for high SNR let say >1000 , 1st sample estimator got reduced noise. But as for as variance is concerned , the sample mean has low variance for high SNR as well.