[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
R: [ns] Multistate Error Model-Markov Model
Hi Stefano,
i'm currently working with the Multistate Error Model.
i have avoided your problem using the transition probabilities in m_transmx.
By using this mechanism the sojurn times are approximately exponentially
distribuited(i.e. memory less).
Alfredo
-----Messaggio originale-----
Da: Stefano Stellin <[email protected]>
A: [email protected] <[email protected]>
Data: domenica 15 luglio 2001 1.40
Oggetto: [ns] Multistate Error Model-Markov Model
>Hi ns-users,
>
>I'm using the Multistate Error Model with two states for modeling the
Markov
>Model and my script runs. But the sojourn times in the two states is
>deterministic, instead the Markov Model requests they are exponentially
>distributed. I understand it is a simplification, but is it possible, and
how,
>to obtain stocastic distribution of sojourn times with the Multistate? I
have
>also thinked to use ErrorModel/Expo but how to give it the error rates? Do
you
>have any suggestion about these problems?
>
>Thanks for your help
>Stefano
>
>