matlab awgn函式加入高斯白噪聲-有解無憂

信噪比

信噪比，電子設備或者通信系統中有效信號和噪聲的比值，英文名稱叫做SNR或S/N（SIGNAL-NOISE RATIO)；

awgn描述

awgn函式可以將白色高斯噪聲添加到信號中，

語法

y = awgn(x,snr) 
y = awgn(x,snr,sigpower) 
y = awgn(x,snr,'measured') 
y = awgn(x,snr,sigpower,state) 
y = awgn(x,snr,'measured',state) 
y = awgn(...,powertype)

描述

y = awgn(x,snr)將白高斯噪聲添加到向量信號x中，標量snr指定了每一個采樣點信號與噪聲的比率，單位為dB，如果x是復數的，awgn將會添加復數噪聲，這個語法假設x的能量是0dBW，
y = awgn(x,snr,sigpower)和上面的語法相同，除了sigpower是x的能量，單位為dBW，
y = awgn(x,snr,‘measured’)和y = awgn(x,snr)是相同的，除了agwn在添加噪聲之前測量了x的能量，
y = awgn(x,snr,sigpower,state)和y = awgn(x,snr,sigpower)是相同的，除了awgn首先重置了正態亂數產生器randn的狀態為整數狀態，
y = awgn(x,snr,‘measured’,state)和y = awgn(x,snr,‘measured’)是相同的，除了awgn首先重置了正態亂數產生器randn的狀態為整數狀態，
y = awgn(…,powertype)和前面的語法相同，除了字串powertype指定了snr和sigpower的單位，powertype的選擇有’db’ and ‘linear’，如果powertype是’db’，那么snr是按照dB為單位測量的，sigpower是按照dBW為單位測量的，如果powertype是線性的，snr是按照一個比率測量的，sigpower是以瓦特為單位測量的，Relationship Among SNR, Es/N0, and Eb/N0
對于SNR和其他的噪聲相對能量測量的關系，查看Describing the Noise Level of an AWGN Channel，

例子

%	Example 1: 
         % To specify the power of X to be 0 dBW and add noise to produce
         % an SNR of 10dB, use:
         X = sqrt(2)*sin(0:pi/8:6*pi);
         Y = awgn(X,10,0);
 
 %	Example 2: 
         % To specify the power of X to be 3 Watts and add noise to
         % produce a linear SNR of 4, use:
         X = sqrt(2)*sin(0:pi/8:6*pi);
         Y = awgn(X,4,3,'linear');
 
 %   Example 3: 
         % To cause awgn to measure the power of X and add noise to
         % produce a linear SNR of 4, use:
         X = sqrt(2)*sin(0:pi/8:6*pi);
         Y = awgn(X,4,'measured','linear');
 
 %   Example 4: 
         % To specify the power of X to be 0 dBW, add noise to produce
         % an SNR of 10dB, and utilize a local random stream, use:
         S = RandStream('mt19937ar','seed',5489);
         X = sqrt(2)*sin(0:pi/8:6*pi);
         Y = awgn(X,10,0,S);
 
%    Example 5: 
         % To specify the power of X to be 0 dBW, add noise to produce
         % an SNR of 10dB, and produce reproducible results, use:
         reset(RandStream.getGlobalStream)
         X = sqrt(2)*sin(0:pi/8:6*pi);
         Y = awgn(X,10,0,S);

matlab 程式正弦+白噪聲


f1=50;  	%  頻率
fs=1000;  	%  采樣頻率
Ts=1/fs;  	%  采樣間隔
N=200;    	%  采樣點數
n=1:N;
y=sin(2*pi*f1*n*Ts);
snr=30;		%  信噪比
r=awgn(y,snr);
subplot(2,1,1);
plot(y,'black')
axis([-inf,inf,-2,2,])
title('原信號');
subplot(2,1,2);
%hold on
plot(r,'black')
axis([-inf,inf,-2,2,])
title('原信號+高斯噪聲');

SNR=10

SNR=20

SNR=30

matlab 程式鋸齒波+噪聲

t = 0:.1:20;
x = sawtooth(t); % Create sawtooth signal.
snr=10;

y = awgn(x,snr,'measured'); % Add white Gaussian noise.
subplot(1,3,1);
plot(t,x,t,y) % Plot both signals.
legend('Original signal','Signal with AWGN');
axis([-inf,inf,-1.5,1.5,])
title('SNR=10');

snr=20;
y = awgn(x,snr,'measured'); % Add white Gaussian noise.
subplot(1,3,2);
plot(t,x,t,y) % Plot both signals.
legend('Original signal','Signal with AWGN');
axis([-inf,inf,-1.5,1.5,])
title('SNR=20');

snr=30;
y = awgn(x,snr,'measured'); % Add white Gaussian noise.
subplot(1,3,3);
plot(t,x,t,y) % Plot both signals.
legend('Original signal','Signal with AWGN');
axis([-inf,inf,-1.5,1.5,])
title('SNR=30');