This thesis demonstrates both analytical and statistical tools for robust design by applying them to a high impedance bootstrapped differential amplifier. This circuit example has only one free variable for optimization. Having a single free variable allows using optimization tools without requiring a volume for each analytical and statitstical method. The thesis is written in tutorial style and includes significant detail. The robust design tools applied are: - Analytical and Statistical Optimization - Analytical and Statistical Sensitivity Analysis - Analytical and Statistical Worst Case Analysis - Expectation Theory and Monte Carlo Distribution Analysis Other techniques that are used are: - Simplification by only retaining significant contributors. - Simplification by use of symmetry. - Short form notation for handling circuit noise contributors. This thesis is intended as a method sampler so that the reader can get acquainted with a wide range of tools, all applied to the same example, and all contained in one reference. Hopefully this will entice the reader to try a new method instead of only the familiar one. And finally the reader will then have a fuller toolbox better equipped to pick the right tool for the right task.
Library of Congress Subject Headings
Electronic circuit design--Mathematical models; Electronic circuit design--Data processing; Engineering design; Control theory
Department, Program, or Center
Electrical Engineering (KGCOE)
Hossenlopp, James, "Methods for robust designs: applied to a high impedance bootstrapped differential amplifier" (1995). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus