Optimization-Oriented Design of RF/Microwave Circuits Using Inverse-Linear-Input Neuro-Fuzzy-Output Space Mapping With Two Different Dimensionality Simulators

In this paper, an efficient optimization technique aligning three-dimensional (3-D) electromagnetic simulator responses with two-dimensional (2-D) electromagnetic simulator responses around space mapped solution, for the electromagnetic design of radiofrequency and microwave 3-D circuits is presented. The interface between 2-D and 3-D electromagnetic simulators is obtained from an inverse linear input space-mapping approach and an output modeling process based on a fuzzy logic technique. It is shown that the technique provides a highly accurate estimation of the 3-D design parameter space of the 3-D radiofrequency and microwave circuits to be designed, for initially fixed design specifications. Designs of first-order, second-order and sixth-order C-band 3-D evanescent rectangular waveguide bandpass filters with dielectric posts show the performance of our approach.