Weird time domain signal

[Adelcio Marques de Souza]

Hello, everyone.
I’m working on a self-seeding system where the RSOA noise is turn into an optical carrier by circulating it on an external cavity. Basically, I have the following elements: RSOA, convert to noise beans, time delay, gaussian optical filter and optical mirror. And it works: after 100 iterations, I get a quasi-CW signal (see Fig. attached). However, there are some “negative peaks” or “dips” in the time domain (see Fig. attached). This is not supposed to happen in a physical system. It is clearly a numerical calculation problem.
Any ideas on what could it be and how can I avoid this?
Thank you very much!

Attachments:

1. 

   ss_time_domain.jpg
   

[Ahmad Atieh]

Dear Adelcio,
I agree with you about the numerical simulation effects. You may use larger number of samples per bit to improve the resolution. You may control the bandwidth of the gaussian filter.
We could suggest better options if we can access your project layout. You may send the project as a zip file to my email at ahmad.atieh@optiwave.com and I’ll take a look.
regards,
Ahmad

[Author]

Posts