RBF_Gradient.m

说明：  模式识别matlab工具箱。一般常见的分类方法都包含在内，还有svm分类。(Pattern recognition matlab toolbox. General common classification methods are included, as well as svm classification.)

高斯勒让德公式，利用高斯勒让德两点公式计算积分，(Chrysler let Germany formulas, Chrysler let Germany two o' clock integral formula)

说明：  希尔伯特黄变换，亲测有效，可放心使用，只是三维图运行会卡顿（三维图部分不确定正确）(Hilbert Huang transform, effective in pro test, can be used at ease, but the operation of 3D map will be stuck (the part of 3D map is uncertain))

这里面有两个程序，两个都是采用4QAM调制。不过个人觉得第一个更容易懂，因为写的很有条理。我改的那个程序就是根据这个来的。 不过在这里面有个问题就是信道系数只能是由高到低，大家可以试试，改成[0.227 0.46 0.688 0.46 0.227]这个是老师给我的仿真中常用的信道系数，这个成为归一化信道，就是系数平方和为1. 第二个程序没有涉及到信道系数，但是这个程序当中，在128个子载波中，前后有32个空载波，这个在OFDM中也是经常用的，这样可以降低误码率。 (And there are two programs, two are based on 4QAM modulation. But personally feel that the first is easier, because the writing is organized. I changed that procedure is based on the past. However, there is a problem in that there is only channel coefficients from high to low, we can try and change into [0.227 0.46 0.688 0.46 0.227] This is the teacher gave me the simulation channel coefficient commonly used, this became normalized channel, is coefficient of the square and is 1. The second program is not related to channel coefficients, but this program which, in the 128 sub-carriers, the before and after the 32 air carriers, the in OFDM is also often used, which can reduce the error rate.)

matlab software bpsk

说明：  otsu自适应阈值分割 用于图像分割二值化(otsu adaptive thresholding segmentation for image binarization)

齿轮啮合刚度曲线,用来计算齿轮啮合动力学作为参考的一个很不错的例子。(It is a similar method,you can change some one from it.)

北邮研究生信息论，许文俊老师，编程作业。离散信道容量的迭代实现。(Beijing University of Posts and Telecommunications graduate information theory, Xu Wenjun teacher, programming assignments. Iterative discrete channel capacity to achieve.)

ldpc_encoder 自己改过的LDPC码编码，貌似是8160,7136.好用的(ldpc_encoder MATLAB 8160,7136)

也是利用最小二乘法对椭圆进行数据拟合，不过考虑到图像大小问题，这里只选取100个边缘点，点的个数也可以自行设置。(Ellipse is the least square method of data fitting, but taking into account the size of the image, here select only edge points 100, the number of points you can set your own.)