A Statistical Approach to Analyze the Impact of Random Discrete Dopant (RDD) and Oxide Thickness Variation (OTV) on the Performance of a Low Power Folded Cascode OTA

The proposed work endeavours to present a statistical study to determine the nature of variation of the major performance parameter i.e. voltage gain of a low power Folded Cascode Operational Transconductance Amplifier (OTA) subject to variations of the two major process parameters—threshold voltage (V<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</inf>) and gate oxide thickness (t<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ox</inf>). The process parameter variations are considered to follow a normal distribution about a mean value with a 3% standard deviation. The OTA circuit is simulated in Spice environment using Python script to generate the entire data set for V<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</inf> and t<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ox</inf> variations. The OTA gain from the generated data set is analyzed statistically using R programming to determine its nature of variation and is compared with three standard distributions: normal, lognormal and Weibull. The study reveals that the nature of variation of OTA gain due to V<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</inf> variation follows a normal distribution where OTA gain due to t<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ox</inf> variation follows a Weibull distribution.