multivariate random numbers generation

How to generate multivariate random number normally distributed in simulink? The mvnrnd generates two different set of vectors with mean=[0 0] variance=[0.001 0;0 0.001] and seed[ 0 0]; whereas in simulink random number generator block generates two random number vectors which are repeatable with same seed.

Answer 1

[Akshat Dalal]

Hi Devyani, you can achieve this in two very simple ways:

1. You can set the value of the "Seed" parameter in the block dialog to be “randi(5)”. This function returns a pseudorandom scalar integer between 1 to "imax", here "imax" is 5. This way each time the model is simulated, the seed value will be initialized to a random integer. Note that I have used the “randi()” function as an example, you may explore other available function as per your requirement.
2. Another option is to use a variable in the base or model workspace as the seed value and then update its value by adding a “StartFcn” callback on the given block. This callback function will execute at the start of every simulation, so you again get a random value for each simulation.