Abstract:High-performance passive devices play a critical role in the radio frequency front-end of wireless systems. Accurately characterizing the electromagnetic (EM) responses of these devices poses a formidable challenge, particularly in high-frequency design. Commercial numerical methods often demand substantial computational resources and necessitate complete recalculation for any structural modifications. This paper proposes an input skip-connections feedforward neural network (ISFNN) for the parametric modeling of passive devices. The ISFNN architecture incorporates multiple skip connections within the layer blocks, which periodically integrate the input design variables into the intermediate hidden layers. This design facilitates feature combination and enhances the extraction capability of design variables. The intermediate layers contain both the original input features and the learned feature information from preceding layers, enabling the model to effectively and robustly capture the nonlinear relationships between the design variables and EM responses. Additionally, a systematic algorithm is proposed to develop and train the ISFNN model. The ISFNN offers a unified solution for both single-physics EM analysis and EM-centric multi-physics (MP) analysis. Compared to other ANN-based models, the ISFNN achieves smaller testing errors with fewer training samples for single-physics EM modeling. Furthermore, in certain applications, conducting MP analysis is more aligned with the actual operating conditions of high-performance microwave components. The nongeometrical parameters are incorporated into the input design features. The ISFNN accurately predicts EM responses using only MP training data, without requiring additional single-physics EM data, transfer functions, or multiple mapping modules. Validation examples demonstrate the better modeling accuracy and efficiency of the ISFNN. Within the design space, the trained ISFNN can generate new MP data in just 0.02 seconds per calculation, significantly reducing time costs while maintaining high modeling precision.

Radial Basis Function Neural Network With Hidden Node Interconnection Scheme for Thinned Array Modeling

Fourier Subspace-Based Deep Learning Method for Inverse Design of Frequency Selective Surface

Modeling of EMI Filter by Means of Recurrent Neural Network with Encoding Layer

A Novel Nonlinear Dynamic System Modeling Approach Using Radial Basis Function Neural Networks

An Equivalent Dipole Model Hybrid with Artificial Neural Network for Electromagnetic Interference Prediction

ROM-Based Real-Time Analysis of Electromagnetic Performance for APAA in a Digital Twin System

2D quasi-periodic array modeling using reduced basis method

An Equivalent Radiation Source Based on Artificial Neural Network for EMI Prediction

Quasi-Periodic Array Modeling Using Reduced Basis with H-2-Matrix Algorithm

Quasi-Periodic Array Modeling Using Reduced Basis Method

ISFNN: an Enhanced Neural Network for Parametric Modeling of Passive Devices with Input Skip-Connections

Hierarchical Attention-Based Machine Learning Model for Radiation Prediction of WB-BGA Package

Radiation Source Reconstruction Based on Artificial Neural Network for Radio Frequency Interference (RFI) Analysis in Complex System

Pls Algorithm For Radial Basis Function Networks

Neural network EM-Field based modeling for 3d substructure in finite element method

Artificial Neural Networks for Microwave Computer-Aided Design: The State of the Art

Analog design of a radial basis function network

Neural Network Modeling for the Reduction of Scattering Grating Lobes of Arrays

A Spatial SEM-Based Shallow Neural Network for Electromagnetic Inverse Source Modeling

Quasi-periodic Array Modeling Using Reduced Basis from Elemental Array

Memory Efficient DGTD Method Based on Radial Basis Function Neural Network Prediction Scheme