genzong

语音分离的研究在语音通信、声学目标检测、声音信号增强等方面有着重要的理论意义和实用价值。而将语音分离技术应用到智能机器人中。让机器人具有智能的听觉，实现声源定位和分离，确定说话人个数，进行人机对话等方面更具有广阔的应用前景。 (Separation of speech in voice communications, acoustic target detection, sound signal enhancement, and so has important theoretical significance and practical value. The voice separation technology applied to intelligent robots. Intelligent robot to the hearing, to achieve sound localization and separation, to determine the number of speaker, the man-machine dialogues in a more broad application prospects.)

采样频率 fs=10000 轴承外环故障信号 fid=fopen( bearingout.dat , r ) 故障 N=1024 xdata=fread(fid,N, int16 ) fclose(fid) xdata=(xdata-mean(xdata))/std(xdata,1) 时域波形 figure(1) plot(1:N,xdata) xlabel( 时间 t/n ) ylabel( 电压 V/v ) db10小波进行4层分解 一维小波分解 [c,l] = wavedec(xdata,4, db10 ) 重构第1～4层细节信号 d4 = wrcoef( d ,c,l, db10 ,4) d3 = wrcoef( d ,c,l, db10 ,3) d2 = wrcoef( d ,c,l, db10 ,2) d1 = wrcoef( d ,c,l, db10 ,1) ( Sampling frequency fs = 10000 bearing outer ring fault signal fid = fopen (' bearingout.dat' , ' r' ) failure N = 1024 xdata = fread (fid, N, ' int16' ) fclose (fid ) xdata = (xdata-mean (xdata))/std (xdata, 1) time-domain waveform figure (1) plot (1: N, xdata) xlabel (' Time t/n' ) ylabel ( ' voltage V/v' ) db10 wavelet decomposition 4 layer one-dimensional wavelet decomposition [c, l] = wavedec (xdata, 4, ' db10' ) 1 ~ 4 reconstructed detail signal d4 = wrcoef (' d' , c, l, ' db10' , 4) d3 = wrcoef (' d' , c, l, ' db10' , 3) d2 = wrcoef (' d' , c, l, ' db10' , 2) d1 = wrcoef (' d' , c, l, ' db10' , 1) )

matlab环境下的adhoc仿真模拟实例(matlab the adhoc simulation under instance)

一个带有GUI、Simulink 3D模拟的四旋翼仿真程序。控制参数、目标点可调。使用时先打开simulink模型，再运行GUI_Config.m即可(With a GUI, Simulink 3D simulation of four-rotor simulation program. Control parameters, the target point is adjustable. First open the simulink model, when used, can run GUI_Config.m)

this compress program can be extracted to a text file that can be copied to mse.m. a matlab program that calculate the mean square eeror for video in the format of yuv such as the test sequence

说明：  在求解每日零息债券的基础之上，计算每日远期利率曲线 (basing on Solving zero-coupon bonds in the daily basis, calculated daily initial forward rate curve in future)

pattern recognition classifiers in Matlab code

Dijkstra算法,关于 Dijkstra算法解决voronoi图的问题,用 Dijkstra算法解决voronoi图中求解最短路径的时候,有一个"dijkstra.m"的文件(Dijkstra algorithm voronoi diagram on Dijkstra algorithm to solve the problem, using Dijkstra algorithm to solve voronoi diagram for solving the shortest path, there is a " dijkstra.m" file)

说明：  利用最小二乘法的方法和3次插值的方法拟合曲线(Curve fitting by least square method and interpolation)

matlab code regarding annealing and computing