DCO Sigma-Delta Modulation's intrinsic problem of linearity
I've got a question on DCO's sigma-delta modulation. The modulator itself is out of the question. However, since the DAC is accomplished inside DCO through discrete small capacitors' on/off, the high intrinsic non-linearity there will destroy the spectrum.
It is impossible to make those capacitor tuning gain linear in my opinion.
I do not know how people solve that problem. Please suggest, thanks.
Reference: JSSCC?2005 Nov. Robert Bogdan "a digitally controlled oscillator in a 90nm digital cmos process for mobile phones"
What intrinsic non-linearity are you speaking of? The cap's are switched between two values to modulate the frequency
Dave
www.keystoneradio.com
What I meant is that if we make control code as thermo-coded, thus any increase of code was represented as turning on one more varactor.
Thus the speed of DCO should decrease linearly, however, we knows that it is impossible. There comes the non-linearity.
Ok. Now I understand. That's the same problem you face in an analog VCO with a nonlinear tuning curve. Read some more of his papers, he's published a lot of the details of their DCO approach, and you'll see how they deal with that non-linearity.
Dave
Would you like to give me 1-2 paper title on that issue?
Which author is the one "his" you mentioned?
Well, you referenced a paper by Staszewski in the start of this thread, that's who I'm referring to as "he". Take a look at IEEE Xplore, there have been many papers published by Dr. Staszewski on that topic.
Dave