tadeoman
Junior Member level 3
- Joined
- Jun 8, 2011
- Messages
- 27
- Helped
- 12
- Reputation
- 24
- Reaction score
- 12
- Trophy points
- 1,283
- Location
- Atlanta, GA
- Activity points
- 1,469
Hello,
I want to ask you EDA-board gurus for suggestions with HSIM Co-simulation and Monte-Carlo simulations.
I want to run a MonteCarlo simulation in a relatively large transistor-level circuit (imagine a microcontroller), but the caveat is that the environment to this circuit is very complex and event-driven.
For example: if signal 'out' is high then I put some input stimuli on the inputs otherwise I use a
different vector stimuli.
This is usually no problem because I can run the VCS-Hsim co-simulation, then the circuit runs in HSIM and the environment runs on Verilog or on a C library. Unfortunately, the cosimulation does not allow to do MonteCarlo simulations, and I really want to model parametric variations at low voltage operation.
I tried looking at VCD files, but AFAIK I can not do an event-driven input, I can only do a time-driven vector inputs. Another option I thought was to rewrite the netlist and add a "sigma" process variation
to each transistor using a given distribution, but I don't know how to do that.
Any suggestions would be greatly appreciated.
Thank you!
I want to ask you EDA-board gurus for suggestions with HSIM Co-simulation and Monte-Carlo simulations.
I want to run a MonteCarlo simulation in a relatively large transistor-level circuit (imagine a microcontroller), but the caveat is that the environment to this circuit is very complex and event-driven.
For example: if signal 'out' is high then I put some input stimuli on the inputs otherwise I use a
different vector stimuli.
This is usually no problem because I can run the VCS-Hsim co-simulation, then the circuit runs in HSIM and the environment runs on Verilog or on a C library. Unfortunately, the cosimulation does not allow to do MonteCarlo simulations, and I really want to model parametric variations at low voltage operation.
I tried looking at VCD files, but AFAIK I can not do an event-driven input, I can only do a time-driven vector inputs. Another option I thought was to rewrite the netlist and add a "sigma" process variation
to each transistor using a given distribution, but I don't know how to do that.
Any suggestions would be greatly appreciated.
Thank you!
