natspline

实现故障诊断！供大家参考!有需要的朋友可以下载过来参考(gu zhang zhengduan )

单载波频域均衡系统中信道估计错误对整个系统的影响(Impact of Channel Estimation Errors on SC-FDE Systems)

基于压缩传感的匹配追踪重建算法研究。给出了OMP的一种改进方案。OMP算法本身耗时过长速度过慢，本文的改 进方案将图像进行分块后再处理，从而大大降低了OMP算法每次迭代的矩阵规模。 实验结果表明，该方案在不明显降低重建效果的同时提高了运算速度。(Matching track reconstruction algorithm based on compressed sensing)

处理超声设备用线阵探头采集到的RF数据，对其进行分析，并进行成像。(signal processing)

A dynamic estimation of channel is necessary before the transmission of the signals since the wireless channel is time–varying..we use MMSE method for channel estimation. The MMSE method calculates the following, MMSE channel estimator is designed to minimize the estimation MSE. The MMSE estimate the channel responses.

特征降维方法，其中包含有典型的特征组合方式——PCA和特征选择中的SFFS，SBS，SFS。共四类常用的特征降维方式。(different features dimensionality reduction methods.)

Matlab source code related to image. which is useful for the developers

SIMPACK，车辆系统动力很好的讲稿，主要讲得建模过程(SIMPACK, vehicle system dynamics very good script, put the main modeling)

说明：  求解大规模无约束最小化问题，使用信赖域算法，其中信赖域子问题使用截断共轭梯度法(The trust region algorithm is used to solve the large-scale unconstrained minimization problem, in which the truncated conjugate gradient method is used for the trust region subproblem)

说明：  分支定界法（branch and bound）是一种求解整数规划问题的最常用算法。这种方法不但可以求解纯整数规划，还可以求解混合整数规划问题。分支定界法是一种搜索与迭代的方法，选择不同的分支变量和子问题进行分支。 对于两个变量的整数规划问题，使用网格的方法有时更为简单。 [1] 通常，把全部可行解空间反复地分割为越来越小的子集，称为分支；并且对每个子集内的解集计算一个目标下界（对于最小值问题），这称为定界。在每次分枝后，凡是界限超出已知可行解集目标值的那些子集不再进一步分枝，这样，许多子集可不予考虑，这称剪枝。这就是分枝定界法的主要思路。(Branch and bound method is one of the most commonly used algorithms to solve integer programming problems. This method can solve not only pure integer programming but also mixed integer programming. Branch and bound algorithm is a method of search and iteration, which selects branches with different branch variables and sub problems. For integer programming problems with two variables, the grid method is sometimes simpler[ 1] Generally, all feasible solution spaces are divided into smaller and smaller subsets repeatedly, which is called bifurcation; And a lower bound of the objective (for the minimum value problem) is calculated for the solution set in each subset, which is called the bound. After each branch, those subsets whose bounds exceed the objective value of the known feasible solution set will not be further branched, so many subsets can not be considered, which is called pruning. This is the main idea of branch and bound method.)