Matlab adds noise / disturbance to data

Normally distributed noise 、 Gaussian noise ：

Generate a dimension of Custom_size, The mean for Mean, The variance of Var The noise of ：

Noise = Mean + sqrt(Var)*randn(Custom_size);

among ,randn()  Returns a from the standard normal distribution （ The mean for 0, The variance of 1） The resulting random scalar .

example 1：

K>> Test = 3 + sqrt(0.5)*randn(1,102400); R = [mean(Test),sum((Test(1,:)-mean(Test)).^2)/(length(Test)-1)]

R =

           3.0017565021829         0.501380264755088

example 2：

K>> Test = -7 + sqrt(0.01)*randn(1,102400); R = [mean(Test),sum((Test(1,:)-mean(Test)).^2)/(length(Test)-1)]

R =

         -7.00017236519557        0.0100363870325614

To raw data RawData Plus the relative error level is Error_level The noise of ：

NoiseData = RawData + Error_level*randn(size(RawData));

example ：

The relative error level added to the data generated by the sine function is 0.1 The noise of ：

%  Defined function f(x)
f = @(x) sin(x);

%  Generate discrete points 
x1 = 0:1/100:2*pi;

F_Noise = f(x1) + 0.1*randn(size(x1)).*f(x1);

figure
plot(x1,F_Noise)

The relative error level here 0.1 It is also the standard deviation above , namely sqrt(Var)

Reference from / Recommended reading ：

https://blog.csdn.net/xiaoyang591/article/details/118544118

Noise category and Matlab Add noise _qq_45479499 The blog of -CSDN Blog _matlab noise