[Gokul Boro .Participant]

Hi Duy Le Nguyen, maximun possible subcarriers and no. of fft points of new OFDM modulator are same.

I would like to suggest you not to use OS12 blocks, because it sometimes gives wrong result.

Did you mean out of 1024 possible subcarriers and 512 subcarriers were used?

If it is so, see the method of bandwidth calculation given in the component library that can be found in help. It is given as

OFDM symbol period = Symbol period * 128 = 128/(Symbol rate) = 2 * 128 /Global bit rate.

Are working on the sample files?

Regards.

[Gokul Boro .Participant]

Thank you Damian.
Can I use WDM analyzer to measure the Sideband power? I was trying to do so but couldnot measure the sideband power separately. I tried setting the parameters lower and upper frequency limit of the WDM analyzer such that only the sideband comes withing that limit. The figure of the spectrum is attached. Resoultion bandwidth of WDM analyzerwas set to be 0.05.

Thank you

Attachments:

1. 

   Untitled.jpg
   

[Gokul Boro .Participant]

Hi Damian, thanks for the reply. But Optisystem 14 is already in already there in the website. Is is just 14?

[Gokul Boro .Participant]

Hi Nguyen, I am unable to open the file. I would like to know why you are using OS12 component? I have contacted the optiwave helpdesk once, where they replied that OFDM 12 block may sometimes give wrong results. Why don’t you use OS13 block instead.

While changing the parameters, did you increase the fiber length?

Tell me the rest of the parameters you are trying in your system, if you are working with DD-OOFDM system.

Regards

[Gokul Boro .Participant]

Hi Nguyen,can you please tell if the BER before insertion of CP was zero?

[Gokul Boro .Participant]

Thank you Damian, for making it clear. I was getting worried about this.

[Gokul Boro .Participant]

Hi Damian,
Thank you for the reply.
I have re-attached the screenshot that you have attached above after editing the picture, where I showed my point of confusion.
Since the EVM measurement shown by the OFDM block was never a zero in any of the iterations, the graph should not have been plot with points in the x-asis (power plot). But there are some points in the x-axis. Theoretically, for this system, the EVM should have been high at lower power and it should have been decreasing as the power is increased and then again EVM increases at a point where the fiber non-linearity come into play.
Please help.

Attachments:

1. 

   evm.jpg
   

[Gokul Boro .Participant]

Thank you Rajguru, but the links you have attached are the ones that were started by myself. If you have any newer linkk, please share it.

[Gokul Boro .Participant]

May be I was wrong. I think

delta_f = Global_bit_rate*/((log2(M)*Max_Possible_Subcarrier)

[Gokul Boro .Participant]

Mr. Singh, BER Test set is a 2-in-1 block. It act as PRBS Generator as well as BER calculator. This block has special processing design for the OFDM Systems.
This block ignores the some errors in bits to calculate the BER. The bits ignored are basically that does not contain data transmitted.
You can go through the component details i the help section of the component.

Regards.

[Gokul Boro .Participant]

Hi Rajguru,
Thanks for replying.
I am aware of those pilots and training symbols. I once somewhere saw that the new OFDM component is completely re-written. Moreover in the simulation files I have attached, these things doesnot seem to be problem because as seen in the OS12 component the BER rate for transmission distance of 100KM the BER was 0, same is the case with OS13 block. But what difference I saw was in the RF spectrum analyzer outputs at both the designs. Do you have any answer for that?

[Gokul Boro .Participant]

Hi Duy Le Nguyen,
I am explaining it based on some papers I went through. In the attached files, the theoretical bandwidth of the OFDM signal is 5 GHz. In order to avoid the Signal-Signal beat interference (SSBI) in the photodiode at the receiver, we have to generate a gap of 5GHz(equal to the bandwidth of the OFDM signal itself) between the Optical OFDM Signal and the main optical carrier. Since we have to generate a gap of 5GHz, we need to center the electrical OFDM signal at 7.5 GHz (calculated as 1.5*bandwidth of OFDM) so that the useful bandwidth of OFDM signal does not get affected by the SSBI noise. Now, about LPF what I can say is that 0.61*bitrate=3.1GHz. If you see at the spectrum analyzer, you can see that the useful BW is within 3.1GHz. So to shape the pulse we use these filters as pulse shaping filters. This is what I have understood. If you think that there is some problem with my explanations (because I am still very weak in communication engg.) you can point it out. Your effort will be appreciated.

Now if I am clear, can you please help me with my problems stated above?
Thank you

[Gokul Boro .Participant]

I am attaching the other design here, because I couldnot attach it in the topic.

Attachments:

1. Direct-Detection-Optical-OFDM-System.osd

[Gokul Boro .Participant]

Hi Ankita,
If you are working on DD-OOFDM then here I got some ideas for you to maximize Q factor. The Q factor is dependent on Launch Power and Guard Band. You can try changing these parameters. But since minimum GB required is equal to bandwidth of OFDM channel, you can better ignore changing GB and just change the Launch power.
Regards

[Gokul Boro .Participant]

Hi Debanjan,
Can you please tell why do you need S/P converter?