frequency separation in dual tone optical signal

[Dr. Dhiman Kakati]

Hi all,
I want to transmit a dual tone or dual wavelength optical signal (i.e. the multiplexed one) through a standard single mode fiber, but when I transmit the signal it gets distorted because of various reasons such as dispersion, phase noise etc. so it would be very much helpful if anyone share his experience working in this field. Thannking you,

Regards,
Dhiman kakati

[Dr. Dhiman Kakati]

is it dependent on bitrate of the system if so what minimum frequency separation will be beneficial for 20 Gbps bit rate.

[Naazira Badar]

Hi Dhiman

Are you using a WDM MUX?

Regards

[Dr. Dhiman Kakati]

Hi Nazira, thank you for the response. though I mentioned about the wavelength multiplexing presently I am not using WDM multiplexure here in my case. I just created the dual tone signal in the electrical domain using two sine wave having different frequency for radio over fiber technology as in the attached snapshot and then applied to the Mach zehnder modulator to have the dual wavelength in the optical domain. please have a look into it.

Regards

Attachments:

1. 

   1111111111111111111.jpg
   

[FAYIQA NAQSHBANDI]

Hi Dhiman

Yes your system performance does depend on the bit rate.If you want your system to support a high capacity of 20 Gb/s you might need to use amplifiers in your design.
If you are using a WDM MUX and DEMUX in your design.I suggest you to check the power level of your signal at the MUX and DEMUX outputs.You can use a spectrum analyser for the same.Depending on the drop you see at the output of the MUX you can set the required gain of your amplifiers accordingly.

Regards

[Dr. Dhiman Kakati]

Hi Fayiqa,

Please refer to the above reply #30903 where I have attached the snapshot of the system of my design.

I will definitely look into these parameters such as power level at the MUX and demux outputs as you have suggested and use amplifier to compensate for the signal attenuation. Thanks for the valuable hint.

Regards,
Dhiman

[alistu]

Hi Dhiman,

Can you please upload your OptiSystem design implementation if it is actually possible, this way, it could be easier to analyze the system and manipulate it if necessary to see if some changes had better be made to it and also to see how it can depend on the bit rate and so on. Thank you.

Regards

[Dr. Dhiman Kakati]

Hi Alistu,
I am attaching herewith the design file, please have a look, I need to have a clear spectrum in optical domain so for this you are allowed to do any kind of modification. you can see the electrical frequency is quite clear in this context, i.e. dual wavelength.

Regards,
Dhiman

Attachments:

1. 11111.osd

[alistu]

Hi Dhiman,

Since the signal has not passed through the fiber yet, dispersion or other channel impairments do not affect the optical spectrum. I reduced the laser linewidth to 0.1 MHz (which is quite consistent with the commercially available low power lasers), and I believe the spectrum got obviously better. Please see if that is acceptable.

Regards

[Dr. Dhiman Kakati]

Hi Alistu,
Thanks for the reply, you can see at the modulator output there are 5 noticeable peak in terms of wavelength. so clearly i need two peaks in optical spectrum also. I have attached herewith the snapshot of the same.

Regards,
Dhiman

Attachments:

1. 

   optical-peak.jpg
   

[alistu]

Both in the screen shot you have attached and using RF spectrum analyzer in the implementation, more than two peaks are clearly visible. And since the signal is up-converted to the pass band, it is not strange to see more peaks. Besides, I put nonlinear effects of laser to zero and the optical spectrum just got more clear, keeping the same peaks. I suggest you evaluate the performance regarding BER to make it easier.

Regards

[Dr. Dhiman Kakati]

Hi Alistu,I am not getting the output as you have explained, please attach herewith the modified design file, that would be very much helpful. Yhanking you.

Regards

[alistu]

I am merely referring to RF spectrum analyzer (the one that has been used after the combiner). Or even the electrical signal you have attached in your former comment clearly shows what seems to be only two peaks consist of more small peaks that are close to each other. The only change I made to your design was to change the linewidth the way I mentioned.

[Dr. Dhiman Kakati]

Hi Alistu,
You are correct that there are two peaks in each frequency of the RF spectrum analyzer which is noticeable, is there any way to make one peak in each frequency of the Electrical domain.
and I made the linewidth which I did not change before (forgot to do so). Thanking you.

Regards,
Dhiman

[alistu]

You’re welcome. The signal is modulated with a frequency of 450 MHz by the quadrature modulator. So when it is up-converted again, two peaks form around the 10 GHz and also around 20 GHz new carrier frequency, each having a 450 MHz distance from the center carrier frequency. This is according to theory and nothing seems to be wrong about it.

Regards

[Dr. Dhiman Kakati]

Thanks alistu for making me clear, I tried making higher frquency such as 10 GHz then the separation increases now I can sure that this is because of the quadrature modulator

Regards,
Dhiman

[alistu]

You’re welcome. And I generally believe it would be better to consider parameters like BER rather than the spectrum as a criteria to decide about system performance, since sometimes the theoretical aspect might be less obvious. Nonetheless, optical spectrum analyzer provides valuable insight into the system performance.

Regards

[Dr. Dhiman Kakati]

Hi Alistu,
definitely i will go for designing the receiver section, and to look for BER, Q-factor and other performance parameters, Thanks for the reply..

Regards,
Dhiman

[alistu]

You’re welcome. I was unaware that the receiver was not implemented, though.

Regards

[Author]

Posts