Artificial Neural Networks for High-Frequency Design
Description
This book explores artificial neural networks (ANNs) and their applications in high-frequency engineering, with particular attention to areas such as antenna design, microwave engineering, and microwave photonics. Emphasis is placed on recent advancements in leveraging ANNs for enhanced forward and inverse modeling, local and global optimization, multi-objective design, uncertainty quantification, and design automation, including system synthesis and unsupervised design approaches. The authors discuss a wide range of ANN architectures, including MLPs, RNNs, LSTMs, ResNets, CNNs, graph neural networks, physics-informed neural networks, neuro space mapping techniques, and multi-fidelity models. The book features detailed case studies and benchmarking of ANN techniques against state-of-the-art methods, offering both practical insight and theoretical grounding.
More details
Persons
Anna Pietrenko-Dabrowska received the M.Sc. and Ph.D. degrees in electronic engineering from Gdansk University of Technology, Poland, in 1998 and 2007, respectively. Currently, she is a Full Professor with Gdansk University of Technology, Poland. She has served as the Associate Editor of Int. J. Num. Modeling, Int. J. Ant. Propag., and Electronics. She has been a guest co-editor of several special issues ISI-ranked journals including Int. J. Num. Modeling, Electronics, and Applied Sciences. She has been a program committee member of international conferences (IEEE MTT-s Int. Conf. Num. EM and Multiphysics Modeling and Optim., Int. Conf. Comp. Science, IEEE Int. Symp. Ant. Propag.). She is a co-author of the monographs Response feature technology for high-frequency electronics. Optimization, modeling, and design automation (Springer, 2024), and Performance-driven surrogate modeling of high-frequency structures (Springer, 2020), and the Editor of Surrogate modeling for high-frequency design: recent advances (World Scientific, 2022). Her research interests include simulation-driven design, design optimization, experiment design, control theory, modeling of microwave and antenna structures, numerical analysis. Since 2021, she consistently made the list of top 2% scientists by Stanford University in the field of Networking & Telecommunications. She published over 250 peer-reviewed research papers. Her h-index is 31 with over 3600 citations.
Content
Artificial Intelligence in Microwave and Antenna Systems: Enabling Smarter Workflows.- Artificial Intelligence Enabled Real Time Failure Recovery of Microstrip Patch Antenna Arrays.- Parametric Modeling of Electromagnetic Behaviors Using Neuro-TF Techniques for High-Frequency Design.- EM Design Optimization using Neuro-TF Surrogates for High-Frequency Design.- Multi-Fidelity Forward Electromagnetic Modeling and Neural Networks.- Multi-Fidelity Neural and Surrogate Models for Electromagnetic Inverse Problems and Digital Twins.- Bayesian Neural Network-Assisted Antenna Design Exploration Technique and Applications.- A Physics-Informed Deep Operator Network for 3-D Time-Domain Electromagnetic Modeling.- Automated Neural-Network-Based Model Generation Algorithms for Microwave Applications.- Consensus Deep Neural Networks, Ensembles of DNNs and Mixture of Experts for Antenna Design.- Modeling of Antenna Electrical and Field Characteristics Using Transformer-Inspired Convolutional Neural Networks.- Digital Predistortion Topologies with Neural Network-Assisted Coefficient Estimators for Highly Efficient Power Amplifiers.- Accurate Data-Driven Antenna Modeling Using Recurrent Neural Networks and Dimensionality Reduction.- Autonomous Deep Learning Frameworks for High-Fidelity Behavioral Modeling of Microwave Transistors.- Artificial Neural Network-Assisted Multi-Surrogate Optimization Framework for High-Frequency Antenna Design.- Machine Learning Strategy for Modeling Dielectric Dispersive Permittivity.- AI-Driven Design for Millimeter-Wave Antenna Arrays.