# c=(exp((-z​^2)/(2*sig​ma^2)))+mn​oise

1 view (last 30 days)
Ajay Joseph on 1 Feb 2014
Commented: Ajay Joseph on 1 Feb 2014
How do you add mnoise which is a 251 data with normal distribution and derivation 0.02. When I added mnoise It says there is an error with + matrices

Amit on 1 Feb 2014
mnoise = 0.02*randn(251,1);
Amit on 1 Feb 2014
Here, the issue is
z = linspace(-7,7,251); % This is a row vector
mnoise = 0.02*randn(1,251); % Now we made this a row vector as well
This time this will work :)
Ajay Joseph on 1 Feb 2014
Damn thanks man The simplest things makes a big difference

John D'Errico on 1 Feb 2014
It appears you have not realized there is a difference between row and column vectors. This is a common mistake, since a vector is just a vector, right? In fact, MATLAB sees a vector as just an array with the row or column dimension equal to 1. That makes a lot of sense in a linear algebra context, which is really where MATLAB comes from.
LINSPACE (like COLON) creates row vectors, so an array of size 1 by n.
A good way to add noise to an array or vector without bothering to know the size is...
y = y + randn(size(y)).*0.02;
RANDN by default generates Gaussian deviates with mean zero and standard deviation 1. Multiplying by 0.02 scales them to have standard deviation 0.02.
Ajay Joseph on 1 Feb 2014
thanks but I guess Amit answer works aswell