Desktop

遗传算法工具箱 给出遗传算法中所有能用到的所有函数，给使用者带来方便。(Genetic Algorithm Toolbox Genetic Algorithm give all to all the functions can be used to bring convenience to users.)

Example of MUSIC function with overlap consideration. Best results are obtained with a big overlap.

光滑比的检验，用于灰色预测模型的建立过程，如果数据达不到要求，就不能用来预测(to exam it whether smooth or not)

this package can simulate a rayleigh multipath fading channel .by running the ,exam.m, the channel will be simulated.,siglfade can simulate a simple rayleigh channel . multipath.m use delay and siglefade to simulate a multiath fading channel.

吉布斯定轨法。 输入：轨道上任意三个点的位矢(R1,R2,R3)； 输出：质点在R2端点时的速度V2；ierr是个判断符号，返回的ierr是零的话，就说明‘R1,R2,R3’共面，如果ierr是1的话说明‘R1,R2,R3’不共面(正常情 况下，要构成轨道的话，‘R1,R2,R3’一定共面 ，即ierr是零，如果ierr是1的话说明输入的‘R1,R2,R3’不共面不能构成轨道。 此函数用于：任意给定三个空间共面的向量，可以确定质点在R2处的速度V2，也就得到了轨道上的一个状态向量(R2,V2),调用前面的函数就可以确定其轨道，及任意时刻的状态。(Gibbs set the track method. Input: track any three points of the bit vector (R1, R2, R3) output: particle at R2 endpoint speed V2, ierr is a judgment symbol, The ierr that is returned is zero, then, it means' of R1, R2,, R3, ' coplanar, if ierr is 1, then description' of R1, R2,, R3, ' not a total surface (under normal circumstances, you want to constitute a track, then,' of R1, R2, R3, ' a certain total surface, ierr is zero, if ierr is an instructions to enter the words ' of R1, R2, R3' coplanarity can not constitute a track. This function is used: for any given three space vector of the surface, can determine the particle at the speed V2, R2 also track a state vector (R2, V2), call the previous function can determine its orbit, and at any time of the state.)

说明：  一个动画模拟的粒子寻找最小值的一个简单程序，有助于学习粒子群法(an animated simulation of particles searching for the minima of a simple function，which helps you learn PSO. )

three phase 120 mode of operation of voltage source inverter

在最佳位置和合同双层办法 分布式发电的径向分布定价 使用混合整数线性规划系统(Bilevel approach for optimal location and contract pricing of distributed generation in radial distribution systems using mixed-integer linear programming)

Mfcc 提取语音特征参数 可以分辨出不同情感(Mfcc speech feature extraction can distinguish between different emotions)

Multiple kernel learning is a model to merge multiple kernels by linear combination. Mostly solving the models are slow due to explicit computation of kernels. Here, we propose to approximate kernel map function explicitly in finite dimensional space. Then, we use dual coordinate descent to solve the SVM. By storing the solutions in primal, we do not have to compute the kernel explicitly. A group lasso regularization on kernel weights is solved with SVM alternatingly. This is a side-project in my research projects with Dr. Yi-Ren Yeh and Dr. Frank Wang in Academia Sinica.