A Robust Bayesian Optimization Framework for Microwave Circuit Design under Uncertainty

In modern electronics, there are many inevitable uncertainties and variations of design parameters that have a profound effect on the performance device. These are, among others, induced by manufacturing tolerances, assembling inaccuracies, material diversities, machining errors, etc. This prompts wide interests in enhanced optimization algorithms take these uncertainty sources into account able to find robust designs, i.e., designs insensitive early cycle. this work, novel machine learning-based framework accounts for is presented. achieved using modified version expected improvement criterion. Moreover, data-efficient Bayesian Optimization leveraged limit number simulations required solution. Two suitable application examples validate robustness significantly improved compared standard methods.