Adaptive Neuro-Fuzzy Models for Conventional Coplanar Waveguides

Abstract

Abstract-In this work a new method based on the adaptive neuro-fuzzy inference system (ANFIS) was successfully introduced to determine the characteristic parameters, effective permittivities and characteristic impedances, of conventional coplanar waveguides. The ANFIS has the advantages of expert knowledge of fuzzy inference system and learning capability of neural networks. A hybrid-learning algorithm, which combines least-square method and backpropagation algorithm, is used to identify the parameters of ANFIS. There are very good agreement between the results of ANFIS models, experimental works, conformal mapping technique, spectral domain approach and a commercial electromagnetic simulator, MMICTL.

Concurrent Neuro-Fuzzy Systems for Resonant Frequency Computation of Rectangular, Circular, and Triangular Microstrip Antennas

“…In [63], the optimal values for the premise and consequent parameters of ANFIS were obtained by the hybrid learning (HL) algorithm [64,65], and the ANN was trained with bayesian regularization (BR) algorithm [66]. In previous works [67][68][69][70][71][72], we successfully also used ANNs and ANFISs for computing accurately the various parameters of the transmission lines and for target tracking.…”

Soft computing techniques on multiresonant antenna synthesis and analysis

“…An equivalent lumped element circuit for MWCNTs-based IDC describes well the measured behavior up to 3 GHz. Thanks to the extracted equivalent circuit, we demonstrate that the capacitive couplings between MWCNTs dominate over nonlinear effects above 10 Since their discovery, carbon nanotubes (CNTs) are emerging as novel nanomaterial for large variety of applications, particularly as electronic material owing to their mechanical, thermal, and unique electrical properties [1]. Electrical properties of individual CNTs, either single-walled (SWCNTs) or multiwalled (MWCNTs) CNTs have been theoretically and experimentally studied.…”

A Generalized Design Procedure for a Microwave Amplifier: A Typical Application Example

“…In this work, the Reflectance Based Method is employed, whose essentials are based upon the simplified real frequency technique given in details [13], to build up the scattering parameters of the Darlington two-ports to match the given generator Γ G and load Γ r terminations to the required source Γ S and load Γ L terminations of the transistor, respectively, for the desired (F , V i , G T ) triplets, as given in the Figs. 2-3.…”

“…In fact, nowadays these two typical nonlinear learning machines have found wide-range applications in the electromagnetic engineering. Typically, neural networks are applied to forming quasi-static modeling for multilayer cylindrical coplanar lines [10], calculation of the impedance of air-suspended trapezoidal and rectangular shaped microshield lines [11], non uniform antenna array synthesis [12], design of the coplanar waveguides combining fuzzy systems [13], passive dipole arrays with together genetic algorithm [14]. On the other-hand, support vectors have become a strong competent method to the neural networks by the typical applications on the linear- [15] and non linear modeling [16] of mesfets; modeling of the microwave devices such as microstrip antenna based on the experimental data [17], mim capacitors [18].…”