matlab简单分析短时傅里叶变换STFT

华中科技大学《数字信号分析理论实践》 学习总结记录

Fs = 5120.0;dt = 1/Fs;N = 32768;df = 500.0/N;T = dt * N;t = linspace(0,T,N);for i = 1:8192x(i) = sin(2*pi*200*i*dt);x(i+8192) = sin(2*pi*400*i*dt);x(i+16384) = sin(2*pi*600*i*dt);x(i+24576) = sin(2*pi*800*i*dt);endZ = spectrogram(x,1024,512);P = 20*log10(sqrt(Z.*conj(Z)));X = linspace(0,Fs/2,513);Y = linspace(0,T,63);mesh(X,Y,P');xlabel('频率')ylabel('时间')view(15,70)

调幅信号谱阵分析

Fs = 5120.0;dt = 1/Fs;N = 32768;df = 500.0/N;T = dt * N;t = linspace(0,T,N);for i = 1:Nx(i) = sin(2*pi*0.3*i*dt)*sin(2*pi*(500+df*i)*i*dt);endplot(x)figureZ = spectrogram(x,1024,512);P = 20*log10(sqrt(Z.*conj(Z)));S = size(P);% 513 = 1024/2+1X = linspace(0,Fs/2,513);Y = linspace(0,T,S(2));mesh(X,Y,P');xlabel('频率')ylabel('时间')view(15,70)

扫频信号谱阵分析

Fs = 5120.0;dt = 1/Fs;N = 32768;df = 500.0/N;T = dt * N;t = linspace(0,T,N);for i = 1:Nx(i) = sin(2*pi*(500+df*i)*i*dt);endplot(x)figureZ = spectrogram(x,1024,512);P = 20*log10(sqrt(Z.*conj(Z)));S = size(P);% 513 = 1024/2+1X = linspace(0,Fs/2,513);Y = linspace(0,T,S(2));mesh(X,Y,P');xlabel('频率')ylabel('时间')view(15,70)

扫频信号生成chirp()

Fs = 1000;T = 0:0.001:12;X = chirp(T,1,12,400,'q');subplot(211)plot(T,X);subplot(212)spectrogram(X,512,256,512,Fs)

频率混叠

Fs = 1000;T = 0:0.001:12;X = chirp(T,1,12,900,'q');spectrogram(X,512,256,512,Fs)