Spice export for INTERCONNECT, netlist trimming about siepic_ebeam_pdk

[lukasc-ubc]

Suggested approach:

References on Graph connected components:

• https://breakingcode.wordpress.com/2013/04/08/finding-connected-components-in-a-graph/
• http://www.cs.cornell.edu/courses/cs2112/2012sp/lectures/lec24/lec24-12sp.html

Starting point:

• in SiEPIC_EBeam_functions:
  • def generate_short_Spice_file(topcell, optical_waveguides, optical_components):
  • def connected_component(i,j):

Approach:

1. Take all components and waveguides, and merge them into a new data structure, say "objects", and include their connectivity. Make it compatible with the [breakingcode].(https://breakingcode.wordpress.com/2013/04/08/finding-connected-components-in-a-graph/) approach.
   • Loop through all components & waveguides, and add to "objects" dataset.
   • use the "connected_component" function (not finished) to identify the neighbours for each object.
2. Modify the breakingcode code to return only objects connected to the laser
3. Create a spice netlist, as a variant of "generate_Spice_file".

[lukasc-ubc]

Order doesn't matter. They just need to end up in the (short) spice export.
The main purpose is to remove the components that aren't part of the
simulation in order to speed up the time consuming import in interconnect.

On Apr 23, 2016, at 12:02 AM, Jaspreet Jhoja [email protected]
wrote:

Do we just want a group of elements or they need to be in a particular
order?

for eg.
Group #1 lukasc-ubc#1: a, b,
c, d, e, f

Can it be like:
Group #1 lukasc-ubc#1: c, b,
a, d, f, e

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
https://github.com/lukasc-ubc/SiEPIC_EBeam_PDK/issues/56#issuecomment-213678168

[jaspreetj]

working on it. Almost finished it.

[lukasc-ubc]

Hi Jaspreet,

Thanks for your pull request! There was a merge conflict, so I just downloaded your file and merged it myself. Yes, please do add this to the MC Simulations.

Do we need the disconnected ports on the top of the device? (I don't think so...)

[jaspreetj]

I just implemented this to the MC simulations and cleared the unused ports.

Do we still need the generate_short_Spice_file(topcell, optical_waveguides, optical_components) function in the SiEPIC Functions?

If not, can I remove the function and add the newly created function to the Functions file?

[lukasc-ubc]

I think it is working well.

I have merged it into the main function, and removed the incomplete function you mentioned.

thank you!
Lukas