channel

lDPC校验矩阵的环检测。包括4环检测算法，6环检测算法及8环检测算法。还包括一个准循环LDPC奇偶校验矩阵的生成算法。(this algrithm can search cycles in LDPC parity check matrix。It contain cycle-4 cycle-6 cycle-8 test algrithm.It also contains a algrithm for generate the LDPC quasi-cyclic parity check matrix )

扩展卡尔曼与粒子滤波结合的程序,用扩展卡尔曼作为提议分布，然后利用粒子滤波进行滤波跟踪，效果比常用的粒子滤波要好！均方根误差也比较小。(Extended Kalman particle filter in conjunction with a program, as proposed by the extended Kalman distribution and particle filter for filtering track, the effect is better than the commonly used particle filter! RMS error is relatively small.)

简单的自己写的matlab例程，适合初学者使用下载快速掌握图像处理。(Matlab write their own simple routines for beginners download quickly master image processing.)

对图像进行纹理分析，分几个方向进行的处理。matlab程序(right image texture analysis, in several directions at the moment. Matlab procedures)

this is new approach for fuel cell control.it is really helpfull.

job shop scheduling for single machine

GMDH的MATLAB源代码，是主程序，也就是gmdhbuild.文件。与之配套的测试文件gmdhtest.m、输出误差gmdhpredice.m会接着发出。(GMDH The MATLAB source code, is the main program, which is gmdhbuild. File. Accompanying test file gmdhtest.m, output error will be followed by a gmdhpredice.m.)

1998年, Watts和Strogatz 提出了小世界网络这一概念,并建立了WS模型。 实证结果表明,大多数的真实网络都具有小世界特性(较小的最短路径) 和聚类特性(较大的聚类系数) 。 WS小世界模型构造算法 1、从规则图开始：考虑一个含有N个点的最近邻耦合网络，它们围成一个环，其中每个节点都与它左右相邻的各K/2节点相连，K是偶数。 2、随机化重连：以概率p随机地从新连接网络中的每个边，即将边的一个端点保持不变，而另一个端点取为网络中随机选择的一个节点。其中规定，任意两个不同的节点之间至多只能有一条边，并且每一个节点都不能有边与自身相连。 (In 1998, Watts and Strogatz proposed the concept of the small world network, and the establishment of the WS model. The empirical results show that most real networks have small world (the smaller the shortest path) and clustering properties (high clustering coefficient). WS small-world model construction algorithm 1, from the rules FIG: consider a nearest neighbor coupling network with N points, they are surrounded by a ring, wherein each node and its left and right adjacent to each of K/2 connected to the node, K is an even number. The 2 randomized reconnection: randomly with probability p from the new connection on each side of the network, coming section of one of the endpoints remain unchanged, while the other endpoint is taken as the random selection of a node in the network. Which states that between any two different nodes can have at most one edge, and each node can not be connected to its own side.)

Finding the center of Omni-directional image

说明：  约束优化问题算法的 matlab 程序，包括： minRosen.m---Rosen 梯度投影法 minFactor.m----乘子法 minconPS.m-----坐标轮换法 minPF.m----外点罚函数法 minMixFun.m-----混合罚函数法 minJSMixFun.m----加速混合罚函数法(Constrained optimization algorithm matlab procedures, including: minRosen.m-- Rosen gradient projection method minFactor.m---- multiplier method minconPS.m----- coordinate rotation method minPF.m---- outside the points penalty function method minMixFun.m----- mixed penalty function minJSMixFun.m---- speed up the hybrid penalty function)