face-recognized

非线性回归通用程序,稍加修改即可用于各种非线性回归(General non-linear regression procedures, a little modification can be used for various non-linear regression)

802.15.4 MAC protocol source code

周期图法谱估计需要对信号序列“截断”或加窗处理，其结果是使估计功率谱密度为信号序列真实谱和窗谱的卷积，导致误差的产生。本程序分别设计了周期图法,最大熵功率谱估计和Welch法以及它们的对比,最大熵功率谱估计目的是最大限度地保留截断后丢失的“窗口”以外信号的信息，使估计谱的熵最大。(Periodogram spectrum estimation signal sequence needs to " cut off" or add the window handle, the result is an estimated power spectral density of the signal sequence of the real spectrum and spectral convolution window, resulting in the generation of errors. The procedures were designed periodogram, maximum entropy power spectrum estimation and the Welch method and their comparison, the maximum entropy power spectrum estimation purpose is to maximize the retention of truncated after the loss of the " window" than the information signal, the entropy of the estimated spectrum maximum.)

应用灰色关联方法进行变压器油色谱故障诊断程序(Transformer oil DGA diagnoses program by gray relationship method)

Conduct a digital system ( electronic lock ) using hierarchic methodology . An electronic lock is a device that allows access or opening of a system , as long as the key or combination to enter match with which it is predefined in said lock .

本程序基于一阶AR模型，u(n)=-0.99u(n-1)+v(n)的线性预测。白噪声v(n)方差0.995.FIR滤波器的抽头数为2.遗忘因子0.98.用RLS算法实现u(n)的线性预测。并附有仿真图片。。(This procedure is based on a first-order AR model, u (n) =-0.99u (n-1)+v (n) of the linear prediction. White noise v (n) the number of taps of the the variance 0.995.FIR filter 2. Forgetting factor 0.98. U (n) of the linear prediction using the RLS algorithm. With simulation Pictures. .)

This program alows you to add noise to a source and the remove the noise by filtering.

local binary pattern matlab source code

按照文献《complex networks》中192-193页方法生成的scale free network，有详细注释。此方法有助于理解scale free network的生成原理，并画出了P（k）图，可以自己验证r值是否正确。(According to literature " complex networks" in the 192-193 page method to generate the scale free network, with detailed notes. This approach helps to understand the generation of scale free network theory, and plotted the P (k) diagram, you can verify their r value is correct.)

Designs an FIR filter given a desired frequency response H_des(w). The design is judged by the maximum absolute error (Chebychev norm). This is a convex problem (after sampling it can be formulated as an SOCP). minimize max |H(w) - H_des(w)| for w in [0,pi] ( where H is the frequency response function and variable is h (the filter impulse response). )