How to use Birnbaum-Saunders dist in SImulink

I have a data set from a process that is in a Birbaum-Saunders distribution. Using it in a set of tests in Matlab shows that it will produce the values I need for simulating that process correctly.

I am building this in Simulink, but am at a loss for how to actually use this non-normal distribution in the Simulink model. Can anyone tell me how to do this, or point me in the right direction?

Answer 1

[patrick1704]

Hi there,

if you are not required to fulfill special coding guidelines, the Matlab function block is always an option, with which you can directly include Matlab code: Include MATLAB code in models that generate embeddable C code - Simulink - MathWorks Deutschland

Otherwise, you can probably use uniformly distributed numbers (Generate uniformly distributed random numbers - Simulink - MathWorks Deutschland) and implement something like inverse transform sampling yourself (Inverse transform sampling - Wikipedia) to spread them according to the desired distribution.

[EK]

Well, I am already using Matlab function blocks for several other things, and was going to use one for this too. But I was hoping that there was some way of easily incorporating the distribution I have already generated, rather than recreating the distribution each time the function is called. Same with implementing an inverse transform to generate Birnbaum-Saunders distributed numbers from a normal distribution.

I am probably missing something obvious about using Simulink, as I am very green and inexperienced with it.