GMM模型

说明：  siPLS和biPLS是两种用于近红外光谱特征波段筛选的变量选择方法，可筛选出与兴关注指标密切相关的特征波段，并去除无关和干扰变量，提高模型稳健性(Bipls and sipls are two variable selection methods for screening the characteristic intervals of near-infrared spectra, which are closely related to performance indicators, to improve the robustness of models.)

对于chebyshev法只能对于均匀线阵，本程序实现对任意结构阵列的波束优化(Chebyshev method for only the uniform linear array, the program structure of arbitrary beam array optimization)

采用LSTM算法用python语言实现的信号时间序列预测，可预测信号的占用度(The LSTM algorithm is used to predict the signal time series in python language)

轨迹跟踪卡尔曼滤波建立在线性代数和隐马尔可夫模型(hidden Markovmodel)上(The trajectory tracking Calman filter is based on linear algebra and hidden Markov models (hidden, Markovmodel))

1.数据时随机产生的（产生1000个随机数），进行排序2.每一次数据排序之后，都可以看到排序的移动，交换次数以及排序的时间。 (1. After randomly generated data (generated 1000 random numbers), every sort 2. Sort the data, you can see the sort of move, switching times and sort of time.)

对互谱法被动测距的原理及算法各个环节进行仿真验证：加权最小二乘估计中权系数的选取；相关窗长度的选取；各频段综合方法等等。最后实现了对9000m—15000m距离的测距(Cross-spectral method for passive ranging all aspects of theory and algorithms for simulation: weighted least squares estimation of the right of the selection coefficient related to the length of the selection window an integrated approach to the various bands and so on. Finally realize the right of ranging from 9000m-15000m)

说明：  朴素贝叶斯算法，以函数形式实现，花了一个上午的时间(Naive Bayes algorithm, in order to function the form of implementation, a morning flower time)

很好很实用的信号处理程序。适合学习和科研参考使用。(Very good very useful signal processing. Suitable for learning and scientific research use and reference.)

说明：  用相场法模拟晶粒的长大的一个程序，用c++编写的，(a program simulation the grain growth by Phase field written using c++)

卡尔曼滤波程序用于时间序列预测，结构简单程序易懂(kalman filter for prediction)