How can i add 20% additive guassian noise to each and every element of vector
13 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Fiaz Ahmad
el 14 de Mayo de 2016
Respondida: Bjorn Gustavsson
el 14 de Mayo de 2016
Suppose i have a fixed value vector a=[1.05;0.2561;.3321]). I want to add 20% Gaussian noise to each value of this vector as follows:
a_with_noise=[a]+[20% Gaussian noise]=====> This should give me; a_with_noise=[1.05+noise;0.2561+noise;.3321+noise]
Here [a] should act as the mean of the Gaussian noise distribution.
Thanks
0 comentarios
Respuesta aceptada
Image Analyst
el 14 de Mayo de 2016
There is a function called randn() that you must have overlooked when you were searching for "random" in the help. Use that. It gives noise with zero mean and unit standard deviation so you'll have to scale it:
% Define our signal - a column vector.
a=[1.05;0.2561;.3321]
% Make the spread of the Gaussians be 20% of the a values
sigmas = 0.2 * a % Also a column vector.
% Create the noise values that we'll add to a.
randomNoise = randn(length(a), 1) .* sigmas
% Add noise to a to make an output column vector.
output = a + randomNoise
0 comentarios
Más respuestas (1)
Bjorn Gustavsson
el 14 de Mayo de 2016
If I get you right this should do it:
function A_p_noise = addgaussiannoise(A,N2S)
A_p_noise = A.*(1+N2S*randn(size(A)));
HTH
0 comentarios
Ver también
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!