MATLAB Answers

How to gaussian white noise with mean, variance and autocorrelation function?

71 views (last 30 days)
I want to input a white gaussian noise (ζ_z )in my system with mean = 0, variance, σ= 3,6 and 12*10^-4 and exponential autocorrelation function with correlation time 𝜏 =0.1. Can anyone help me with this?
The equations are given as-


Sign in to comment.

Accepted Answer

Chaitanya Mallela
Chaitanya Mallela on 23 Oct 2020
Edited: Chaitanya Mallela on 23 Oct 2020
The White Gaussian Noise should contain a multivariate random numbers with three different realizations, each realization with mean 0 and variance 3,6,0.0012 respectively.
N=1000; % Input sample length
M = [0 0 0]; % Mean vector
var = [3 6 12*10^-4]; % Variance vector
Cov = var.*eye(3,3); % Covariance matrix
x = mvnrnd(M,Cov,N); % MultiVariate random vector added to the input
x1 = autocorr(x(:,1)); % Space Correlation of three realizations
x2 = autocorr(x(:,2));
x3 = autocorr(x(:,3));
Since time is a separate dimension, we get the Gaussian Random Variables by fixing a particular time instant(say t) in the Gaussian Random Process to obtain the above three realizations. So, we consider the time instants in such a way it satisfies the time step auto correlation function of the Gaussian Random Process.

  1 Comment

Musanna Galib
Musanna Galib on 23 Oct 2020
Thank you very much for the reply. I am not expert in this field. I didn't understand the explanation related to implement auto-correlation time. Can you kindly elaborate how to implement the autocorrelation time. I am actually trying to reproduce some result. They used these parameters mentioned in the question.
This noise will be used to give a force. I am quoting the line - "The stochastic force is an exponentially correlated noise with fixed standard deviation and correlation time 0.1 s with exponential autocorrelation function."

Sign in to comment.

More Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by