matlabcode2

MLP with wieght and multilayer

matlab编写的ISODATA算法，实验数据为Iris，代码注释详细，运行良好！(Matlab prepared by the ISODATA algorithm, the experimental data for the Iris code comment in detail, a good run!)

基于matlab冒泡法排序的实现,Matlab排序函数,在Matlab中排序某个向量（一维）时，可以使用sort.(Bubble Sort )

mimo仿真，适用于初学者，不错的源程序欢迎下载(mimo simulation for beginners, a good source welcome Download)

说明：  粒子滤波代码，是用MATLAB实现的,可以参考(Particle filter code is the use of MATLAB to achieve, can refer to)

Matlab被称为第四代计算机语言，利用其丰富的函数资源，可使编程人员从繁琐的代码中解脱出来。Matlab用更直观、符合人们思维习惯的代码，代替了C和Fortran语言的冗长代码，给用户带来的是最直观、最简介的程序开发环境。(Matlab is called fourth-generation computer language function, programmers can be freed from the tedious code. Matlab with a more intuitive, the code in line with people' s habits of mind, instead of a lengthy code in C and Fortran languages, giving users the most intuitive and Introduction program development environment.)

扩展卡尔曼滤波算法的matlab程序，比较完整。(Extended Kalman filter algorithm matlab program, complete.)

Adaboost是一种迭代算法，其核心思想是针对同一个训练集训练不同的分类器(弱分类器)，然后把这些弱分类器集合起来，构成一个更强的最终分类器（强分类器）。 load clouds [test_targets, E] = lijsada_boost(patterns, targets, patterns, 100, Stumps ,[]) train_patterns 每列为一样本 train_targets 每列为一样本目标 100 :Number Of Iterations Stumps: Weak Learner Type Learner s parameters :[] adaboost +stumps testRightR = 0.7522(Adaboost is an iterative algorithm)

DVB-T 2K signal generation Eq. (2.1.4) vs. 2N-IFFT.DVB-T 2K 信号发生器matlab源代码。(DVB-T 2K signal generation Eq. (2.1.4) vs. 2N-IFFT.DVB-T 2K signal generator matlab source code.)

最小均方（LMS）自适应算法就是一中已期望响应和滤波输出信号之间误差的均方值最小为准的，依据输入信号在迭代过程中估计梯度矢量，并更新权系数以达到最优的自适应迭代算法。LMS算法是一种梯度最速下降方法，其显著的特点是它的简单性。这算法不需要计算相应的相关函数，也不需要进行矩阵运算。(Minimum mean-square (LMS) adaptive algorithm is that one has to respond to the expectations and filtering the output signal of the mean square error between the value of the minimum-based, based on the input signal is estimated in the iterative process of gradient vector, and to update the weights in order to achieve the most Adaptive iterative algorithm gifted. LMS algorithm is a steepest descent gradient method, and its significant features is its simplicity. This algorithm does not require calculation of the corresponding correlation function, nor the need for matrix calculation.)