▍1. ComplexNetworks
说明: 复杂网络MATLAB工具包和源程序,用于计算复杂网络特性的工具包。(Complex network MATLAB toolkit and source code)
说明: 复杂网络MATLAB工具包和源程序,用于计算复杂网络特性的工具包。(Complex network MATLAB toolkit and source code)
地球固连坐标系与地球惯性坐标系的转换矩阵程序,考虑的章动、岁差、自转等。(Earth' s solid even coordinate system with the Earth inertial coordinate system transformation matrix program, consider the nutation, precession, rotation, etc..)
这是一个matlab程序,包括了面波频散计算,及最小二乘反演的导数计算内容。(This is a matlab procedures, including a surface wave dispersion calculation, and calculation of the derivative of least squares inversion content.)
说明: 利用matlab求解方程,方程求解并画图(Equation solving and drawing)
声子晶体能带结构计算程序,对初学者帮助很大,希望能帮助需要的人(People phononic band structure calculation program for beginners great help, hoping to help needed)
用于多学科设计优化中代理模型技术,移动最小二乘法是代理模型方法的一种(For multidisciplinary design optimization in the agent model, the moving least squares method is a kind of agent model)
偏微分方程数值解上机实习 运用古典显式差分格式求解二维扩散方程的初边值问题 运用P-R差分格式求解二维扩散方程的初边值问题
偏微分方程数值解上机实习 运用古典显式差分格式求解二维扩散方程的初边值问题 运用P-R差分格式求解二维扩散方程的初边值问题
基于Matlab产生光纤陀螺随机噪声和信号,利用固定步长和可调步长的LMS自适应滤波、最小二乘法、滑动均值三种方法进行处理,最后用阿兰方差评价降噪效果。(FOG Based on Matlab generate random noise and signal, using a fixed step size and adjustable step size LMS adaptive filtering, least squares method, the sliding average of three methods for processing, and finally with Alan variance evaluation noise reduction.)
说明: 局部线性回归方法及其稳健形式已经被看作一种有效的非参数光滑方法.与流行的核回归方法相比,它有诸多优点,诸如:较高的渐近效率和较强的适应设计能力.另外,局部线性回归能适应几乎所有的回归设计情形却不需要任何边界修正。(Local linear regression methods and their solid form has been seen as an effective non-parametric smoothing method. Contrary to popular kernel regression methods, it has many advantages, such as: higher efficiency and stronger asymptotic adaptation design capacity. In addition, the local linear regression to adjust to the return of the design of almost all cases does not require any boundary amendment.)
阵列处理作业,MVDR或者其他高分辨方法,由于信号源无法确定,造成估计误差,采用对角加载的 增加单位阵法 ,以及 六源准则法 获取优化结果。(Array processing, MVDR, or other high-resolution method, the source can not be determined, resulting in the estimated error, the use of diagonal loading ' to increase the unit matrix method' , and ' six-source criteria method' to obtain optimal results.)
C-N解偏微分方程的程序。解的是定步长的抛物型偏微分方程: du/dx - a * d2u/dx2 = 0 在程序中可以更改 a 的值以实现不同系数的解。在循环中改变 a 的值以实现变系数. 该说明在压缩包是也有(CN solution of partial differential equations procedures. Solution is scheduled to step parabolic partial differential equation: du/dx- a* d2u/dx2 = 0 in the procedure can be changed in order to realize the value of a different coefficient of the solution. Change in the cycle in order to realize the value of a variable coefficient. The note in the compressed packet is also)
三维的有限时域差分源程序,边界条件为pml(Three-dimensional FDTD source, boundary conditions for the PML)
移动最小二乘曲线拟合原理,论文中分析了一维和二维的例子,并且同最小二乘法进行比较。(Moving least squares curve fitting principle, the paper analyzes the one-dimensional and two-dimensional case, and compared with the least squares method.)
说明: 该文件内容用于相空间重构参数时间延迟和嵌入维数的计算,时间延迟的计算包含去偏复自相关函数法和互信息法,嵌入维数的甲酸包含G-P算法和FNN算法。(The contents of this file are used to calculate the time delay and embedding dimension of phase space reconstruction parameters. The calculation of time delay includes depolarization complex autocorrelation function method and mutual information method. The embedding dimension of formic acid includes G-P algorithm and FNN algorithm.)
对于网格上每个点求取法向量的方法,程序实现的比较好(Compute normal vectors of 2.5D triangulation)