CO-OFDM Architecture

[alistu]

In CO-OFDM transmitter, a 90degree phase shift is used (as indicated in the file attached which shows a block diagram of the system). Why has not this phase shift block been used in Optisystem sample file “Coherent Detection Optical OFDM system.osd”?

[alistu]

And here is the block diagram of the CO-OFDM system simulated by Optisystem which does not contain the phase shifter.

Attachments:

1. 

   ofdm.png
   

[Damian Marek]

Couldn’t see your attached pic of the block diagram. Can you try reattaching as a jpg or png?

I may be able to guess at the answer though. Those X couplers used in the OptiSystem design actually cause a 90 degree phase shift when the input is coupled to the opposite output. The bottom branch of the transmitter will have a 90 degree phase shift compared to the upper branch.

[Ravil]

Hi alistu, a phase shift is required block in OFDM to create orth. components. It is included in “Coherent detection” block. In my opinion, it should be placed after MZM.

[Ravil]

Did you try to ‘run’ that scheme, by the way?

[alistu]

Here is the block diagram of the system which was not attached.

Attachments:

1. 

   1.jpeg
   

[alistu]

Yes, I have run it and it works.
There has been used a phase shifter in the coherent detection block, which is according to the theory. But none in the transmitter…

[Damian Marek]

Ok there is actually a 90 degree phase shift in the OptiSystem project layout, but it is incorporated into the X coupler. In the simulation (and also in physical X couplers) the signal that is coupled to the adjacent waveguide experiences a 90 degree phase shift. If you look at the transmitter and receiver you will see X couplers that incorporate this phase shift.

[alistu]

That’s the point I hadn’t noticed. Thanks.

[Ravil]

Thank you for the explanation Damain! Do you know the purpose of including Electrical Gain in one of the MZM arms at the transmitter?

[Damian Marek]

The electrical gain is there to facilitate an efficient dual driven modulator. Each branch does not have to modulate the entire Vpi, only half of it.

In a phase shift keying scheme then both branches would have the same amplitude.

[Ravil]

Did anybody test this scheme for its max transmission capacity? (Just wandering…)

[Damian Marek]

By max transmission capacity, do you mean max bit rate to maintain a certain BER?

[alistu]

If by “capacity” you mean Bitrate-Distance product, I have tested it and to my surprise I could only achieve 10Gbps for 360km, which is much less than that of any other paper I’ve studied about CO-OFDM since 2007 (for example, in Alatawi’s 2013 master’s thesis at University of Denver, a system with 50Gbps for 6000km was achieved using Optisystem v.12 without exploiting any types of equalizers and so on). So I would also like to know if anyone has tested the system in this regard. Thanks in advance.

[alistu]

How is receiver sensitivity considered in Optisystem coherent optical OFDM example discussed above? I believe there should be such a parameter in PIN photodetectors, but there isn’t. And sensitivity is not something that can be neglected. Can anyone help please?

[Damian Marek]

The receiver sensitivity is not explicitly shown or set within the PIN Photodiode, as it depends on the users definition (which Q factor is chosen as the threshold). However, the noise and bandwidth which can be set within the Photodiode changes the sensitivity of the photodiode and a nice little example of this is found at this link:

http://psut.edu.jo/sites/yazan/classes/Spring2010/MicrowaveLab/FExp4.pdf

Let me know if this helps!

[alistu]

Thanks a lot! I got what I needed to know.

[Author]

Posts