pei_gt82

同时辨识模型阶次及参数算法。用阶次递推算法，结合AIC法——利用赤池信息准则辨识上例的模型阶次和参数。(At the same time identify model order and parameters algorithm. By order recursive algorithm, combined with AIC method- the use of Akaike information criterion to identify the cases of the model order and parameters.)

说明：  一种基于概率的数据降维处理方法：Stochastic Neighbor Embedding(Based on the probability of data dimensionality reduction approach: Stochastic Neighbor Embedding)

说明：  经过个人修改的matlab科学计算程序，供初学者使用(MATLAB scientific computing and visualization simulation literature amended program can achieve numerical conversion between)

对于给定某一线性时不变网络（不含受控源），根据网络属性及结构，通过编程实现自动生成其关联矩阵A和一个树。(Given a linear time-invariant network (excluding controlled source), according to the network properties and structure of the automatically generated by programming its associated matrix A and a tree.)

几种计算阶乘、阶乘之和的算法，由于计算机存储问题，一般阶乘不可能算得太大，只是一种方法。(Several computing factorial, factorial, and the algorithm, due to computer storage problems, the general factorial can not be considered too great, is only a way.)

西安电子科技大学葛德彪教授书后所附fortran程序(Xidian University, Professor Ge Debiao book accompanying fortran program)

给出变量的上下边界、初值和代价函数，能够搜索代价函数最小值时的变量取值。属于带约束的优化算法，可以用来求算非线性方程组。(STRSCNE is a Matlab code for constrained nonlinear systems of equations F(x)=0 l<=x<=u where F: R^n--> R^n, l and u are vectors of dimension n. Non-existent lower and upper bounds, i.e. entries of l and u equal to minus o plus infinity, are allowed. The algorithm is a globally convergent procedure that combines Newton method and an elliptical trust-region approach. The elliptical trust-region is defined employing a scaling diagonal matrix D and the trust-region subproblem is approximately solved by the dogleg method. Only strictly feasible iterates are generated. Various input/output options are provided, and we refer to the code itself for further documentation. )

ADAM （Adaptive Moment Estimation）是另外一种自适应学习率算法，它结合动量梯度 下降法，在不同参数方向上采用不同学习率，保留前几次迭代的梯度，能够很好 的适应于稀疏数据。(ADAM (Adaptive Moment Estimation) is another adaptive learning rate algorithm, which combines momentum gradient. The descent method, which uses different learning rates in different parameter directions and retains the gradients of previous iterations, is very good. It is suitable for sparse data.)

说明：  使用最小二乘法拟合，数据在程序里。本质是使用fiminuc函数求最小值。欢迎下载。(Use the least square method to fit the data in the program. The essence is to use fiminuc function to find the minimum value. Welcome to download.)

给出接收信号眼图及系统仿真误码率，可实现对二维数据的聚类，matlab开发工具箱中的支持向量机。( The received signal is given eye and BER simulation systems, Can realize the two-dimensional data clustering, matlab development toolbox support vector machine.)