simulating pulse width variation

[sutapanaki]

Hi,

I have a circuit that basically produces pulses with a pulse width depending on a control voltage. I use spectre to simulate it. I need to simulate the variation of the pulse width w.r.t. power supply variations. My idea was (and I may be wrong here) to perform pss-pac simulation having an ac voltage at the supply. For that I created a veriloga block which I included in the circuit. That block measures the pulse-width and outputs that value as an output voltage. The problem is that if I use last_crossing in combination with @ cross for better accuracy, pss doesn't like it and says there are hidden states. If I use only last_crossing then simulation goes through but the pulse-width the pss extracts is wrong.
I would appreciate if someone can suggest a way to perform that pulse-width measurement.
Thank you in advance.

 

[leo_o2]

Please simulate with transient and measure it with Calculator in cadence.

 

[sutapanaki]

Thank you for the reply.
I have considered your proposal. It s not so easy to do. My pulse width is less than 1ns. If I want to fit at least one period of the sine wave that I put on the supply I'll need a frequency of more than 1GHz. If frequency is smaller, then the pulse will kind of "sample" part of the period.