Overview of Microwave Device Modeling Techniques Based on Machine Learning

This paper overviews the microwave device modeling techniques based on machine learning. The proposed methods are divided into two categories, active and passive components modeling. These modeling techniques aim to characterize the microwave device efficiently and accurately. The active components consist of MESFET, HEMT, pHEMT, and so on. The passive devices include inductor, capacitor, Lange coupler, coplanar waveguide, filter, etc. A variety of modeling techniques has been successfully used, e.g., artificial intelligence, support vector regression, genetic algorithm, etc. Increased investigations published in the literature are a true testament of the appeal based on machine learning.