Hi all,
Thanks for your time. I'm working on a project regarding the T-sensor. To reduce the mismatch of current mirrors and diodes, I employ the DEM technique. It is effective in theory by using it and the mismatch is reduced by 1/root(N). However, how effective is the DEM exactly? And how to simulate it using Virtuoso?
emi
Hi @sprite_emi
This document explains the principle of DEM and how to implement it. In order to simulate DEM in Virtuoso you would have to create either a digital logic (using Verilog/VHDL) and preforming a Mixed-Signal simulation (which will be very slow in my opinion), either create a Verilog-A block that will do the bit shuffle for you.
Hopefully, that helps.