Why getting different Results with Optsystem?

[Taiwo Ambali Abiola]

Hi all,

Pls can you explain to me why we were getting different BER?. I am a demonstrator of Optisystem in my institution and I realized my student were getting different value of BER despite using the same parameters on their systems. If the computer could cause this, would this not jeopardize the validity of Optisystem?. Kindly help on what could be the source of the problem. I have over 30 students and I observed they got different results in the attached assignment given to them.

I asked them to disable the effect of attenuation and dispersion at different interval and observed the effect on the system.

Thanks for your help.

Attachments:

1. Attenuation-and-dispersion.osd

[Ashu verma]

Hello Taiwo,I have seen you system in which you have varies the link distance.first of all why you kept attenuation disable?

As you said it gives different values every time,it may be due to the random data of prbs.you can duisable it from prbd>random seed
then you will get same results.Will you please share the objectives that what actually you want to study.Because every transmission medium has some attenuation ,you cannot disable it.Rather than this you can use amplifier to boost your signal and amplify signal which degraded by fiber..I have attched the modified system,find attachment

Attachments:

1. Attenuation-and-dispersion1.osd

[Damian Marek]

Sam is right! By default each time the simulation is run a random value is used for noise and also things like the bit sequence generator. This can make the resulting BER vary from time to time and from machine to machine. What you can do is use a much longer bit sequence length, which will smooth out the inconsistencies between simulations.

Sort of like how increasing power will reduce shot noise!

[Taiwo Ambali Abiola]

Thanks both Mr. Sam and Mr. Damian, I was only trying to show them the effect of attenuation and dispersion in the optical network. I asked them to disable the dispersion in order to first monitor the effect of attenuation on the received signal and later we disabled attenuation and enabled dispersion so as to observe the influence of dispersion on the network. Later, both were enabled and comparisons were done in terms of the BER. The point I was trying to raise was why would there be variation in the signal received on different machines despite using the same parameters.

I agree with the fact that the randomness of the PRBS might have caused it and perhaps the length of the PRBS sequence. Thanks for your contributions. That means if I want to judge the students base on the level of the received signal, I must change the random seed 0.

[Author]

Posts