fujian1

Deasonalize the data adn agian seasonalize it.

支持向量机是用于分类与回归的技术。由于其自身的诸多缺点，如无法获得概率输出，需要估计一个误 差参数C，以及必须使用Mercer 核函数等。相关向量机算法，克服了SVM 上述缺点，RVM 能获得与SVM 相比拟的推 广性能，并且更为稀疏。在此基础上，文中介绍了一种RVM 回归用于分类的新分类方法，用RVRC 来表示。并通过 实验证明了它的可行性。(Support vector machines for classification and regression techniques. Due to its many shortcomings, such as the probability that the output can not be obtained, need to estimate an error parameter C, and the need to use Mercer kernel function and so on. Relevance vector machine algorithm to overcome these shortcomings of SVM, RVM and SVM can be obtained comparable to promote performance and more sparse. On this basis, the paper introduces a RVM regression was used to classify new classification method, RVRC to represent. And through experiments proved its feasibility.)

100个matlab小程序，主要是一些基本的matlab程序(100 matlab small procedures, mainly some basic matlab program)

说明：  jerk模型，已经封装好。入门请勿下载，原始轨迹数据用txt导入，谨慎下载(Jerk model, already packaged. Please do not download at the beginning. The original track data is imported by txt and downloaded carefully.)

说明：   DC Defect Correction Method. X = DC(A,B) attempts to solve the system of linear equations A*X = B for X. The N-by-N coefficient matrix A must be square and the right hand side column vector B must have length This uses the defect correction method until default convergence criteria or MAXIT iterations is reached.

Boundary Integral Method Code

this is simulation of LTE Phy-layer but it is not complete yet not all the codes i done by my self

说明：  IMPLEMENTATION OF ECONOMICAL DISPATCH

和Huffman-Tree一样，Shannon-Fano coding也是用一棵二叉树对字符进行编码。但在实际操作中呢，Shannon-Fano却没有大用处，这是由于它与Huffman coding相比，编码效率较低的结果（或者说香农-范诺算法的编码平均码字较大）。(And Huffman-Tree, like, Shannon-Fano coding is using a binary character coding. However, in practice it, Shannon-Fano, but not much use, since it is compared with the Huffman coding, coding efficiency is the result of a lower (or Shannon- Fano coding algorithm larger average code word).)

Model Predictive Control Exercise pt 3