Noise bins and MMF

[Abdallah Ismail]

Dear,
I’m designing a system that have Loop Control with 8 and connected to measured index MMf and amplifier and BPF,the amplifier generate noise but the measured index MMf produce errors because the noise bins in an input to it.
I tried to enable “convert noise bins” in the global parameters but the same error still appear.
error message “Invalid Input signal,Noise bins are detected at the input”
can anyone help me to solve the problem.

[Damian Marek]

Hi Abdallah,

I remember this problem from a different question you posted. I don’t remember exactly what your project file looked like, but I remember I fixed it by removing the loop. I believe the problem starts because noise bins are being created in the loop somewhere and are propagated back to the input of the MMF.

I know it is not an elegant solution but I think if you just duplicate the components you want to loop and place them serially the problem will be resolved.

Let me know if this helps

[Abdallah Ismail]

Hi Damian,
I replaced the loop by a series of my link and started only with two stages (200Km) ,but The same problems still exist, it executes for the first length but the second fiber section doesn’t accept noise bins at input.
It is a clear that the measured index MMF doesn’t accept signal with noise bins and not the loop.
I thought that “Convert noise bins ” will not make the noise a separate signal but after test,the same problem still appear.
Waiting your support.
Best Regards,
Abdallah

[Damian Marek]

Could you please attach your project file? I don’t remember exactly what I did.

Thanks

[Abdallah Ismail]

Dear Damain,
I attached both FMF design and Optisystem design,please change the path at first,I expanded the loop and started only with two loop 200km but this error mention above appeared.
by the way, Could you help me also in knowing why i receive BER =0.5 although the constellation is appearing seems like good.
Regards,
Abdallah

[Damian Marek]

Hi Abdallah,

I am looking into this project for you, but it is pretty complex so it might take me some time.

Regards

[Damian Marek]

Hi Abdallah,

Are you using OptiSystem 12? I’m running your file with OptiSystem 13 and it is running fine. Just replaced the DSP with the new Universal DSP component.

The BER was large, because of the long multimode fiber. Do you have reason to believe that the long fiber should perform better?

Regards

[Abdallah Ismail]

Thanks Damian,
I’m using Optisystem 13
I’m not sure if you enabled noise in optical Amplifier or not because i disabled it.
The multimode fiber is working as Few Mode Fiber and supports only 6 modes (LP01,LP02,LP11a,LP11b,LP21a,LP21b) which can work for long distance.
Please enable EDFA Noise and check the MMF noise bins error.
Best Regards,

[Damian Marek]

Hi Aballah,

There seems to be a problem with the way noise is handled and the spatial information. The amplifier block is not properly converting noise bins and so the spatial components are not working properly. I have sent this bug off to the OptiSystem team to investigate and fix. It should be possible to get it into the next release 13.0.2 coming out this fall. You can get around this problem by using White Light source and a power combiner to couple in “noise”.

However, I am still looking at other sources of error in the project file, which may be due to the multimode fiber.

Sorry for the inconvenience!

[Abdallah Ismail]

Thanks Damian for your support,I hope it’ll be fixed in the near future.
Regarding the design, I don’t know how the constellation visualizer produce good results and the BER test sets is giving BER=0.5.
I replaced the M-ary threshold detector with Decision component and enabled “optimize the decision”and found that the BER become zero.
But I feel it is not realistic because it uses a copy of the original data as a reference.
Also Does white noise source will produce noise for all modes or i have to use mode generator after it?
I hope you can help me.
Best Regards,

[Damian Marek]

The white light source should add noise to all modes regardless of having a mode itself. I’ve passed on this problem to the OptiSystem team, I believe the problem is coming from the delay created by the multimode fiber and the DSP component trying to compensate for it.

Regards

[Abdallah Ismail]

Hi Damian,
Is it possible to get a clear constellation diagram and the error became the half of data?.
Do you think that M-ary Threshold and PSK-decoder have timing problem?
Also what do you thank if i used Matlab Component to reproduce data from I&Q component,will it make a difference and calculate Q and hence BER?.

[Damian Marek]

First off, the BER is ~0.5 because the data is being read incorrectly. Even guessing at the bit every time you would be right half the time. On average the worst BER you can get is around 0.5.

The constellation diagram I am getting is rotated, yet still distinct levels. The rotation comes from phase delay of the signal with respect to the local oscillator. So I believe the delay is causing the majority of the problem.

I think the next step is to try setting the delay compensation manually (no DSP) using the Decision component. The multimode fiber report in the component properties will explicitly tell you the delay calculated.

The Matlab component is definitely an option but I’m not certain if it would be easier than trying the step above.

[Abdallah Ismail]

This is the constellation of the recived symbols attached as pictures,from these constellation results,do you think that there is phase delay or error.

[Damian Marek]

Are those constellation diagrams after the DSP? If so they might be corrected wrongly, the DSP was created with the single mode fiber in mind.

[Abdallah Ismail]

Ya Damian ,this is after DSP,I changed the DSP coefficients (dispersion Coefficients and slope )with respect to Multimode fiber.
Also I changed the filter and re-sampling according to the new data rates.
Do you think that it can’t deals with MMf even i changed these coefficients?

[Damian Marek]

I would try first to get the system working without the DSP personally, because I am not a DSP expert, even if this means turning off all sources of noise in the system (chrom. dispersion, loss, etc.) Just leave the mode delay on.

I believe the error is coming from the DSP trying to “fix” the delay of the signal, which manifests as a rotation of the constellation diagram. Phase noise which also comes from the modal dispersion cause the values to spread out in an arc. If the delay is quite large, which can happen in multimode fiber, the rotation might be past 90 degrees and then the DSP is going to have a hard time finding the correct position.

Let me know if this advice gets you anywhere!

Regards

[Abdallah Ismail]

I compressed the files and uploaded it again.

[Author]

Posts