- This topic has 4 replies, 2 voices, and was last updated 8 years, 3 months ago by
shaymaa Tahhan.
-
AuthorPosts
-
-
August 15, 2016 at 9:03 am #41117
shaymaa TahhanParticipantDear all,
I am new to optisystem and till now I have worked on it for 504 hours, so please answer to my question because I am in a big puzzle and have a hard time.
I am trying to perform exactly the same system in the attach file and I am not fully understand all the basics.
1. what is the biasing voltage that I should apply to the modulators? does the first modulator have the same biasing voltage settings of the second modulator ??2. what are the settings of the rise time and the fall time of the NRZ generator in my case?
3. What is the required bandwidth of the multiplexer and demultiplexer and should they be same or not for the same output number of the input channels?
4. what is the noise figure frequency of the EDFA and how much the bandwidth ??
5. Is the Q factor that i got from the BER analyzer in dB unit or unit less? if it is a unit less could I write a script somewhere to draw it in dB?
for now these are the most important question and if you require my to attach the optisystem file I will do.
kind regards,
Shaymaathese setting affect the Q factor
-
August 17, 2016 at 6:43 pm #41128
Ashu vermaParticipantHi shymaa
There is some good questions you have asked. First of all one other forum member has query related to similar topic of DWDM at 25G Hz spacing and 40G data rate. Kindly also check her discussion time to time as you will most probably face similar issues while implementation.
Kindly attach the simulation file if you made roughly and answer of these question as followsKindly check the optisystem samples
Samples>transmitter> duobinary return to zeroThis will show the single channel DRZ format and you can also check the modulator settings of DRZ. As ican see, two modulators are having different modulation voltages however bias voltages are fixed.
Kindly double click on modulator and go to help for more information about component.You can consider rise time 0 to 1. but for your case 0.01 is fine and you will get acceptable results.
EDFA amplifier noise and gain depends upon the pump power wavelength and length of fiber. In general we consider noise 4dB of amplifier EDFA
Q-factor that you can see on BER is in linear domain (dimensionless) However you can convert values manually by by applying log function to it.
Yes you can use the script for conversion, but this will be done by help of optiwave technical assistance.
Hope you get some answers and if you can attach file then it would be better
Regards
-
August 17, 2016 at 7:01 pm #41129Ashu vermaParticipant
Hi, i got a solution of Q (db)
As by default scale of y axis set to Q fctor( dimensioess), you need to change it to logarithmic
Steps to change
run simulation > double click on BER analyzer >on left side a blue box appears as i attached in snapshot > thed clck on it and go to propertirs> y axis> click on it> dialog box appear > change scale type to logarithmic
here you will Q(dB) values from y axis. -
August 17, 2016 at 7:05 pm #41130Ashu vermaParticipant
Find Attachement
-
August 20, 2016 at 12:13 pm #41133shaymaa TahhanParticipant
Dear Sam,
Thank you for your reply. Attached the noise bandwidth of the EDFA that i meant before. thank you for clarification about logarithmic scale.For now I am only have these two problems
1> what is the noise bandwidth that i should put in the EDFA to get the maximum Qfactor ?
2> what is the required bandwidth of demultiplexer and multiplexer to get the maximum Q factor?again please find the attached pdf in which all the description of the simulation that I am working on.
Then i want to replace the DCF by FBG to find whether it can be more efficient with duobinary or not?
so how to calculate the length of the FBG ? how to get the Dispersion ?
for Ideal FBG i now that for a length of SMF which has (16 ps/nm km ) the dispesion will be the multiplexing of these values but in minus sign like for 100Km the FBG dispersion will be -1600ps/nmbut for FBG how?
-
-
AuthorPosts
- You must be logged in to reply to this topic.