problem with using poisson distribution function.

[sohini basu]

I am trying to model a network with Poisson traffic When using exponential(with an arrival rate of 250 packets per second or an interarrival rate of 0.004

When using **.interarrivaltime = exponential(0.004) , I am getting results for my delay measure but when I use **.interarrivaltime = poisson(250) then in that case I am not being able to get any results.


Can anyone point out the reason.

[Rudolf Hornig]

Poisson distribution if the number of events happening in a given fixed timeframe. Why do you set the interArrivalTime to be a poisson distribution? Inter arrival time is as it's name suggest the time between two arriving events.

If you have a large number of independent events then:

the time between the events follows an exponential distribution.

the number of events happening under let's say 1s follows a poisson.

[11187162]

I want to model a network with varying traffic that follows poisson distribution. For example, packets arrival rate is (0.004s to 1s). Can anyone suggest what would be the syntax for "interarrivaltime" for that?

Thank you!

[Khalid Hussain Mohammadani]

 Did you put any time unit with 250? I think without time unit it may not work.

--
You received this message because you are subscribed to the Google Groups "OMNeT++ Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to omnetpp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/omnetpp/10329fa9-86f0-4c43-8f02-830064efdf4bn%40googlegroups.com.

[11187162]

I want to model a network with varying traffic that follows poisson distribution. That means, the packet arrival rate be different, let's say ranging between (0.004s to 1s). Would anyone tell me what should be the syntax for that?

Thank you

[Alfonso Ariza Quintana]

In a discrete simulation, you can set like parameter the arrival time, the number of events per second is a more complex to model, it is possible, this implies a big modification of the code. If you consider that the exponential distribution is the inverse of the Poisson distribution if you want to model a Poisson distribution of value Landa, you can use an exponential distribution to model the interarrival time with a mean of 1/Landa, if you measure the interarrival times in interval of one second, you can see that the number of events follow a Poisson distribution of mean Landa

---

De: omn...@googlegroups.com <omn...@googlegroups.com> en nombre de 11187162 <11187...@gmail.com>
Enviado: jueves, 5 de noviembre de 2020 9:39
Para: OMNeT++ Users <omn...@googlegroups.com>
Asunto: Re: [Omnetpp-l] Re: problem with using poisson distribution function.

 

To view this discussion on the web visit https://groups.google.com/d/msgid/omnetpp/ba011806-de97-4860-97bf-bfa7c50ffea5n%40googlegroups.com.

[11187162]

Hello Alfonso, so with **.interarrivaltime = exponential(0.004s) (where 0.004 is the mean arrival rate), what is the minimum and the maximum arrival rate?

Thank you

[Alfonso Ariza Quintana]

the minimum value for a double is 2.22507e-308, but this value will collapse the simulator, it will generate e308 events per second 

---

De: omn...@googlegroups.com <omn...@googlegroups.com> en nombre de 11187162 <11187...@gmail.com>

Enviado: viernes, 6 de noviembre de 2020 6:40

To view this discussion on the web visit https://groups.google.com/d/msgid/omnetpp/c97b6bef-54fb-4544-a92d-d4ee140361e1n%40googlegroups.com.

[11187162]

Hello Alfonso, is it e308 for all time, irrespective of the value of the mean arrival rate? I meant considering mean arrival rate of 0.004s, is the value e308? If I change mean arrival rate, is it still e308?

I wanted to model that the arrival rate changes over time, not fixed. Any suggestion please? Thank you.

[Alfonso Ariza Quintana]

2.22507e-308 second => 1/2.22507e-308  = 4.494e+307 events per second

0.004s => 1/0.004 = 250 events per second

 

Enviado desde Correo para Windows 10

 

De: 11187162
Enviado: viernes, 6 de noviembre de 2020 13:01
Para: OMNeT++ Users
Asunto: Re: [Omnetpp-l] Re: problem with using poisson distribution function.

 

Hello Alfonso, is it e308 for all time, irrespective of the value of the mean arrival rate? I meant considering mean arrival rate of 0.004s, is the value e308? If I change mean arrival rate, is it still e308?

 

I wanted to model that the arrival rate changes over time, not fixed. Any suggestion please? Thank you.

 

On Friday, November 6, 2020 at 7:25:00 PM UTC+11 Alfonso Ariza Quintana wrote:

the minimum value for a double is 2.22507e-308, but this value will collapse the simulator, it will generate e308 events per second 

 

De: omn...@googlegroups.com <omn...@googlegroups.com> en nombre de 11187162 <11187...@gmail.com>

Enviado: viernes, 6 de noviembre de 2020 6:40

Para: OMNeT++ Users <omn...@googlegroups.com>
Asunto: Re: [Omnetpp-l] Re: problem with using poisson distribution function.

Hello Alfonso, so with **.interarrivaltime = exponential(0.004s) (where 0.004 is the mean arrival rate), what is the minimum and the maximum arrival rate?

 

Thank you

On Thursday, November 5, 2020 at 8:00:27 PM UTC+11 Alfonso Ariza Quintana wrote:

In a discrete simulation, you can set like parameter the arrival time, the number of events per second is a more complex to model, it is possible, this implies a big modification of the code. If you consider that the exponential distribution is the inverse of the Poisson distribution if you want to model a Poisson distribution of value Landa, you can use an exponential distribution to model the interarrival time with a mean of 1/Landa, if you measure the interarrival times in interval of one second, you can see that the number of events follow a Poisson distribution of mean Landa

To view this discussion on the web visit https://groups.google.com/d/msgid/omnetpp/732be216-d52a-4ef5-b289-43df973b21d2n%40googlegroups.com.

 

[11187162]

Hello Alfonso, so with exponential(0.004), on average 250 events occur and this rate is a constant over time; Not a variable. Is it? Thanks Sir.

[Alfonso Ariza Quintana]

The rate is constant in the experiment, you can program several experiments with different values

To view this discussion on the web visit https://groups.google.com/d/msgid/omnetpp/e2e66084-0ce6-4aa1-9880-051b5dce749en%40googlegroups.com.

 

[Rudolf Hornig]

TIP: In fact, you can specify a a changing rate, by using the simTime() function inside ned. If you can derive your rate value from the current simtime: e.g.

arrivalrate = exponential(simTime() < 20s ? 2s : 0.01s) 

is working, meaning that in the first 20s the mean arrival rate if 2s and after that it's 0.01s, however you must be aware that this is evaluated ONLY when the next event is schedules, so rate is not changing at 20s but when the first event has delivered after the 20s mark. This may or may not be a problem for you. Of course any other function would work that can use simTime() as an input.

[11187162]

Thanks for the tips. If I want to change rate multiple times, for example first 20s, the mean arrival rate be 2s, in next 20 second it is 1s, and after that it is 0.5s, what should be the syntax? Would you please suggest me?

Thank you.

[Alfonso Ariza Quintana]

It will depend on what you want. If you want to extract the results for every rate, the best if program different experiments if you want to extract how the network reacts when the rate change, in this case, you need to program a change in simulation time

---

De: omn...@googlegroups.com <omn...@googlegroups.com> en nombre de 11187162 <11187...@gmail.com>

Enviado: miércoles, 11 de noviembre de 2020 0:56

To view this discussion on the web visit https://groups.google.com/d/msgid/omnetpp/0a4a468f-24e6-4987-b797-2c938202b251n%40googlegroups.com.

[Rudolf Hornig]

Nothing prevents you to use the ?: operator multiple times i.e. you can also create an "if else" ladder just like in any language where ?: is available... 

arrivalrate = exponential(simTime() < 20s ? 2s : simTime() < 40s ? 1s : 0.5s)

[11187162]

Thanks a lot!

[11187162]

Is there any way to program poisson arrival rate with auto increments. For example, I found slotDuration = ${gwSlotDuration=0.01..1 step 0.01}s. Similar to this, how i can code so that mean arrival rate will be incremented from exponential(0.01s) to exponential(1s) with step 0.01?

Many thanks.