Artificial neural networks and space mapping for EM-based modeling and design of microwave circuits
Cargando...
Fecha
2013-06
Autores
Rayas-Sánchez, José E.
Título de la revista
ISSN de la revista
Título del volumen
Editor
Springer
Resumen
Descripción
This chapter reviews the intersection of two major CAD technologies for modeling and design of RF and microwave circuits: artificial neural networks (ANNs) and space mapping (SM). A brief introduction to artificial neural networks is first pre-sented, starting from elementary concepts associated to biological neurons. Elec-tromagnetics (EM)-based modeling and design optimization of microwave circuits using artificial neural networks is addressed. The conventional and most widely used neural network approach for RF and microwave design optimization is ex-plained, followed by brief descriptions of typical enhancing techniques, such as decomposition, design of experiments, clusterization and adaptive data sampling. More advanced approaches for ANN-based design exploiting microwave knowledge are briefly reviewed, including the hybrid EM-ANN approach, the pri-or-knowledge input method, and knowledge-based neural networks. Computa-tionally efficient neural space mapping methods for highly accurate EM-based design optimization are surveyed, contrasting different strategies for developing suitable (input and output) neural mappings. A high-level perspective is kept throughout the chapter, emphasizing the main ideas associated with these innova-tive techniques. A tutorial example using commercially available CAD tools is fi-nally presented to illustrate the efficiency of the neural space mapping methods.
Palabras clave
Computer-aided Design (CAD), Design Automation, RF and Microwave Modeling, EM-based Design Optimization, Artificial Neural Networks (ANN), Space Mapping (SM), Nowledge-based Neural Networks (KBNN), Neural Space Mapping
Citación
J. E. Rayas-Sánchez, “Artificial neural networks and space mapping for EM-based modeling and design of microwave circuits,” in Surrogate-Based Modeling and Optimization: Applications in Engineering, S. Koziel and L. Leifsson, Ed., New York, NY: Springer, 2013, ch. 7, pp. 147-169.