Analysis of a Simple Method of CMOS IC Design for Yield Optimization

A simple approach for CMOS integrated circuit (IC) design taking into account a process variability and oriented towards optimization of a parametric yield has been presented. Its concept is based on cumulative distribution functions of random variables representing IC performances subject to process variations. In the method it has been assumed that CMOS process statistical data are expressed in terms of so-called process parameter distributions. Thus the design centering is done via layout parameter tuning. The approach relies on maximizing the probability that random variables corresponding to IC performances remain within the performance boundaries. Also, a methodology for statistical characterization of CMOS process has been briefly described. Finally, the method operation has been illustrated using analytical and SPICE models of CMOS inverter, operational amplifier and ring oscillator.