An accurate wavelet neural-network-based model for electromagnetic optimization of microwave circuits

In this paper, a relevant automated electromagnetic (EM) optimization metho d and a novel, fast, and accurate artificial neural network are proposed fo r the efficient CAD modeling of microwave circuits. We lay the groundwork f or our investigation of radial wavelet neural networks WNNs trained by BFGS (Broyden-Fletcher-Goldfarb-Shanno) and LBFGS (limited memory BFGS) algorit hms and their application to determine the scattering parameters of the cir cuit under study. Wavelet theory may be exploited in deriving a good initia lization for the neural network, and thus improved convergence of the learn ing algorithm. The optimization method combines a rigorous and accurate glo bal EM analysis of the device performed with a finite-element method (FEM) and a fast neural model deduced from its segmented EM analysis. Finally, ex perimental results, which confirm the validity of the WNN model, and good a greement between theoretical optimization results and experimental ones are reported. (C) 1999 John Wiley & Sons, Inc.