[Tech SupportParticipant]

Please see the examples provided in OptiSystem 12 Samples\Advanced modulation systems\QAM systems. We have several examples of 16-QAM systems for long distance.

For 64-QAM, we are planning to include examples in our next release (OptiSystem 13) for June/July 2014.

[Tech SupportParticipant]

Ahmad: Your assessment is correct. The Measured-Index Multimode Fiber and Parabolic-Index Multimode Fiber components are expecting multi-mode spatially defined signals. If you use the Multimode Generator with all the power allocated to the primary mode then you will obtain a single mode laser signal with extra information on the mode field type and size (complex field data in the x-y plane) – this resembles most closely your laser in the lab (since it has a field diameter). This information is used by the Measured-Index Multimode Fiber and Parabolic-Index Multimode Fibers for the overlap integrals (used in the mode solver). The accuracy is thus pretty good. For the Linear Multimode Fiber block there is no need for spatial complex field information as we do not call up a mode solver (only a transfer function is applied).

If you connect a Spatial Receiver or Spatial PIN to the output of the Measured Index or Parabolic Fiber then it will apply the spatial aperture against the modal field envelopes. The resulting truncated field data will then be applied as an attenuation to the time-domain waveform of the signal before being converted to an electrical signal. So the temporal effects associated with the signal (multimode dispersion) are held by the primary signal (I recommend that you try out the View Signal Visualizer – this will show you the baseline signal (complex signal as a function of time) and also the complex field data in the x-y plane for each calculated mode (this data sets are separate)).

[Tech SupportParticipant]

Here’s some additional information from our OptiSystem team:

—————————————————————————————————-
The best way to immediately save data into Matlab is with the Matlab component

We would add a signal fork at the location at which you would like to retrieve the data (connecting one of the forks to the Matlab component).

As an example an m-file can be created that reads in an electrical signal and horizontally concatenates the vector (from each sweep) to an array (thus creating a large signal array representing all the sweeps). The example setup is shown in the enclosed images where a “SweepNum” parameter is first setup in the global parameters (from 1 to 5), followed by the “Amplitude” of the Pulse Generator (which is linked to the SweepNum) and a parameter SweepNum that is created in the Matlab component and linked also to the Global Parameter “SweepNum”.

The m-file that can be used is shown below (and also enclosed).

i = SweepNum;
SignalVector = InputPort1.Sampled.Signal’;

if i == 1
TotalSignal = SignalVector;
else
TotalSignal = horzcat(TotalSignal,SignalVector);
end
—————————————————————————————————-

Attachments:

1. 

   ExampleProject.jpg
   

2. 

   Global.jpg
   

3. 

   Matlab.jpg
   

4. 

   Pulse.jpg
   

1. ReadData.zip

[Tech SupportParticipant]

The optical delay creates a null signal. If the input to the component has, for example, 5 separate signals, then the delay should be set to 5 (in order to create 5 separate null signals).

If the 32 users are combined into a single transmission channel (one signal) then the delay only needs to be set to 1 (the default value).

I suggest running the design with the delay set to 1 at all locations to see if the design converges. Alternatively, all the delays can be removed from the design, and within Global Parameters/Signals the “Initial delay” check box can be selected and the number of “Iterations” can be set to, for example, 10.

[Tech SupportParticipant]

A periodic structure would be most easily created with the VB Scipt.

[Tech SupportParticipant]

OptiBPM does not need to sample frequently in Z. The propagation step is often set quite long, because the structures for which BPM is appropriate vary slowly with Z. The sample you have looks under-sampled in Z.

[Tech SupportParticipant]

Yes, OptiFiber can calculate the modal index of cladding modes.

[Tech SupportParticipant]

There are several examples in the “samples” folder. Have you had a chance to look at these?

[Tech SupportParticipant]

It depends on the details of the design of the Photonic Crystal Fiber. We can simulate some of them, but not all.

[Tech SupportParticipant]

OptiGrating will let you select only one mode for input. You can select as many modes as you like for the simulation. If the grating causes coupling from the input mode to any of those selected modes, the power lost from the input mode to the selected mode will be taken from the transmission spectrum of the input mode. The interaction will be apparent from observing the transmission spectrum of the input mode. The graphics do not permit display of all selected modes. If only two modes are selected, all 4 curves will be shown (forward and back waves of each of the two modes). If more than two modes are selected, only forward and back waves of the input mode are shown.

[Tech SupportParticipant]

That’s awesome!

[Tech SupportParticipant]

It needs to be an NVIDIA CUDA-based GPU card.

[Tech SupportParticipant]

The file was built in 12.2.1, so that would be the issue. Attached is a picture of the design.

Attachments:

1. 

   PolSwitch.jpg
   

[Tech SupportParticipant]

What Windows OS & version of OptiSystem are you using?

[Tech SupportParticipant]

I`ve attached a .zip rather than the .osd. Please try again.