COMPARISON OF ADAPTIVE-NETWORK-BASED FUZZY INFERENCE SYSTEM MODELS FOR ANALYSIS OF CONDUCTOR-BACKED ASYMMETRIC COPLANAR WAVEGUIDES

Abstract

A method based on adaptive-network-based fuzzy infer- ence system (ANFIS) is presented for the analysis of conductor- backed asymmetric coplanar waveguides (CPWs). Four optimization algorithms, hybrid learning, simulated annealing, genetic, and least- squares, are used to determine optimally the design parameters of the ANFIS. The results of ANFIS models are compared with the results of conformal mapping technique, a commercial electromagnetic simula- tor IE3D, and the experimental works realized in this study. There is very good agreement among the results of ANFIS models, quasi-static method, IE3D, and experimental works. The proposed ANFIS models are not only valid for conductor-backed asymmetric CPWs but also valid for conductor-backed symmetric CPWs.