几个FDTD程序，对初学者很重要，很有参考价值。

多抽样率信号处理，最大似然（ML）准则和最大后验概率（MAP）准则，一种噪声辅助数据分析方法,考虑雨衰 阴影 和多径影响，鲁棒性好，性能优越，使用大量的有限元法求解偏微分方程。( Multirate signal processing, Maximum Likelihood (ML) criteria and maximum a posteriori (MAP) criterion, A noise auxiliary data analysis method, Consider shadow rain attenuation and multipath effects Robustness, superior performance, Using a large number of finite element method to solve partial differential equations.)

说明：  GA-SVM预测数据，输入训练样本，运用测试样本测试(Ga-svm prediction data, input training sample, use test sample test)

1、GAmain是这个主函数，需要根据问题修改其中的遗传算法参数和最后的画图部分 2、加入非线性寻优之后效果更好 3、加非线性寻优后需要相应更改nonlinear子函数的各个参数

关于matlab信号分析与处理的相关程序，通过matlab来调用实现-On the matlab signal analysis and deal with the relevant procedures, through matlab to call the realization of

使用LMS算法的噪声消除

matlab 关于association rule 的自己写的函数，有3个文件, association.m：h = association(m, i, j) i=>j, m是数据，h是support和confidence，该函数只适用于单个数据 ass_item: h=ass_itset(m, a, b) 同上，但是可用于多个数据（m为数组） assrule: h = assrule(m, threshold1, threshold2) 该函数用于classification， 得到规则，threshold1为要求的support,threshold2为要求的confidence，h 则为符合要求的规则及其support和confidence，前2列为规则，后2列为其support和confidence (matlab on the association rule to write functions, there are 3 files, association.m: h = association (m, i, j) i => j, m is the data, h is the support and confidence, this function applies only to a single Data ass_item: h = ass_itset (m, a, b) it is the same as above, but it can be used for multiple data (m can be matrix) assrule: h = assrule (m, threshold1, threshold2) the function used for classification,get the rules, threshold1 is the require of support, threshold2 is the required of confidence, h is the rules and their support and confidence, the former two columns as a rule, the latter two columns as one of its support and confidence)

标准线性卡尔曼滤波程序，没加蒙特卡罗实验，可参考《雷达数据处理》一书(linear Kalman filtering)

盲源分离中的二阶盲辨识方法识别模态参数~该程序只适用于正定和超定的程序，对于欠定情况并不适用。

new crosstalk algorithm using matlab

使用WOLA滤波器组的均衡器什么的，网上原来下载的代码不全，补充了缺省的几个函数（也从网上搜来），基本可以运行了(Calibrated parametric equalizer reference implementation using WOLA filterbank,code by Benny S鋖lberg 2010)