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.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Data availability
The datasets generated during the current study are available from the corresponding author on reasonable request.
Code availability
The code used in this study is available from the corresponding author upon reasonable request.
Materials availability
All materials used in this study are available from the corresponding author upon reasonable request.
References
Roy C, Wu K (2022) Homotopy optimization and ann modeling of millimeter-wave siw cruciform coupler. IEEE Trans Microw Theory Tech 70(11):4751–4764
Faraji A, Sadrossadat SA, Jin J, Na W, Feng F, Zhang QJ (2024) Hybrid batch-normalized deep feedforward neural network incorporating polynomial regression for high-dimensional microwave modeling. IEEE Trans Circuits Syst I-Regul Pap 71(3):1245–1258
Yan S, Zhang Y, Liu W, Liu G, Shi W (2022) A novel electromagnetic centric multiphysics parametric modeling approach using neuro-space mapping for microwave passive components. Photon 9(12):960. MDPI
Zhang HX, Huang L, Wang WJ, Zhao ZG, Zhou L, Chen W, Zhou H, Zhan Q, Kolundzija B, Yin WY (2020) Massively parallel electromagnetic-thermal cosimulation of large antenna arrays. IEEE Antennas Wirel Propag Lett 19(9):1551–1555
Yan S, Zhang Y, Jin X, Zhang W, Shi W (2018) Multi-physics parametric modeling of microwave passive components using artificial neural networks. Prog Electromagn Res M 72:79–88
Feng F, Na W, Jin J, Zhang J, Zhang W, Zhang QJ (2022) Artificial neural networks for microwave computer-aided design: The state of the art. IEEE Trans Microw Theory Tech 70(11):4597–4619
Ravelo B (2018) Multiphysics model of microstrip structure under high voltage pulse excitation. IEEE J Multiscale Multiphysics Comput Technol 3:88–96
Xu Z, Ravelo B, Maurice O (2019) Multiphysics tensorial network analysis applied to pcb interconnect fatigue under thermal cycle aggression. IEEE Trans Electromagn Compat 61(4):1253–1260
Akstaller W, Weigel R, Hagelauer A (2020) Electrothermal modeling of surface acoustic wave resonators and filters. IEEE Trans Ultrason Ferroelectr Freq Control 67(11):2423–2432
Zhou Z, Wei Z, Tahir A, Ren J, Yin Y, Pedersen GF, Shen M (2023) A high-quality data acquisition method for machine-learning-based design and analysis of electromagnetic structures. IEEE Trans Microw Theory Tech 71(10):4295–4306
Liu W, Feng F, Zhuo Y, Zhang J, Lin Q, Ma K, Zhang QJ (2022) Electromagnetic parametric modeling using combined neural networks and rlgc-based eigenfunctions for two-port microstrip structures. IEEE Trans Microw Theory Tech 70(11):4666–4682
Wei Z, Zhou Z, Wang P, Ren J, Yin Y, Pedersen GF, Shen M (2023) Automated antenna design via domain knowledge-informed reinforcement learning and imitation learning. IEEE Trans Antennas Propag 71(7):5549–5557
Wu Y, Pan G, Lu D, Yu M (2022) Artificial neural network for dimensionality reduction and its application to microwave filters inverse modeling. IEEE Trans Microw Theory Tech 70(11):4683–4693
Sedaghat M, Trinchero R, Firouzeh ZH, Canavero FG (2022) Compressed machine learning-based inverse model for design optimization of microwave components. IEEE Trans Microw Theory Tech 70(7):3415–3427
Xiao LY, Shao W, Jin FL, Wang BZ, Liu QH (2021) Inverse artificial neural network for multiobjective antenna design. IEEE Trans Antennas Propag 69(10):6651–6659
Kanmaz TB, Ozturk E, Demir HV, Gunduz-Demir C (2023) Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces. Optica 10(10):1373–1382
Feng F, Na W, Jin J, Zhang W, Zhang QJ (2021) Anns for fast parameterized em modeling: The state of the art in machine learning for design automation of passive microwave structures. IEEE Microw Mag 22(10):37–50
Feng F, Zhang C, Ma J, Zhang QJ (2016) Parametric modeling of em behavior of microwave components using combined neural networks and pole-residue-based transfer functions. IEEE Trans Microw Theory Tech 64(1):60–77
Qi S, Sarris CD (2022) Deep neural networks for rapid simulation of planar microwave circuits based on their layouts. IEEE Trans Microw Theory Tech 70(11):4805–4815
Yahya SI, Rezaei A, Nouri L (2021) The use of artificial neural network to design and fabricate one of the most compact microstrip diplexers for broadband l-band and s-band wireless applications. Wirel Netw 27(1):663–676
Xiao LY, Shao W, Jin FL, Wang BZ, Joines WT, Liu QH (2019) Semisupervised radial basis function neural network with an effective sampling strategy. IEEE Trans Microw Theory Tech 68(4):1260–1269
Na W, Yan S, Feng F, Liu W, Zhu L, Zhang QJ (2021) Recent advances in knowledge-based model structure optimization and extrapolation techniques for microwave applications. Int J Numer ModelElectron Netw Device Fields 34(5):2879
Chen J, Zhang QJ, Guo Q, Feng F, Liu W, Zhang J, Jin J, Ma K (2024) Advanced neuro-tf modeling technique incorporating quadratic approximated vector fitting of parametric pole/residue extraction. IEEE Trans Microw Theory Tech 72(2):966–980
Liu ZX, Shao W, Ding X, Peng L, Jiang B (2022) Neural network with fourier series-based transfer functions for filter modeling. IEEE Microw Wirel ComponLett 32(7):823–826
Jin J, Feng F, Zhang J, Yan S, Na W, Zhang Q (2020) A novel deep neural network topology for parametric modeling of passive microwave components. IEEE Access 8:82273–82285
Ma J, Dang S, Li P, Watkins GT, Morris K, Beach M (2023) A learning-based methodology for microwave passive component design. IEEE Trans Microw Theory Tech 71(7):3037–3050
Zhang W, Feng F, Gongal-Reddy VMR, Zhang J, Yan S, Ma J, Zhang QJ (2018) Space mapping approach to electromagnetic centric multiphysics parametric modeling of microwave components. IEEE Trans Microw Theory Tech 66(7):3169–3185
Zhang W, Feng F, Liu W, Yan S, Zhang J, Jin J, Zhang QJ (2021) Advanced parallel space-mapping-based multiphysics optimization for high-power microwave filters. IEEE Trans Microw Theory Tech 69(5):2470–2484
Zhang W, Feng F, Yan S, Na W, Ma J, Zhang QJ (2020) Em-centric multiphysics optimization of microwave components using parallel computational approach. IEEE Trans Microw Theory Tech 68(2):479–489
Na W, Liu W, Zhu L, Feng F, Ma J, Zhang QJ (2018) Advanced extrapolation technique for neural-based microwave modeling and design. IEEE Trans Microw Theory Tech 66(10):4397–4418
Liu YF, Peng L, Shao W (2022) An efficient knowledge-based artificial neural network for the design of circularly polarized 3-d-printed lens antenna. IEEE Trans Antennas Propag 70(7):5007–5014
Cao Y, Wang G, Zhang QJ (2009) A new training approach for parametric modeling of microwave passive components using combined neural networks and transfer functions. IEEE Trans Microw Theory Tech 57(11):2727–2742
Feng F, Na W, Liu W, Yan S, Zhu L, Ma J, Zhang QJ (2019) Multifeature-assisted neuro-transfer function surrogate-based em optimization exploiting trust-region algorithms for microwave filter design. IEEE Trans Microw Theory Tech 68(2):531–542
Feng F, Gongal-Reddy VMR, Zhang C, Ma J, Zhang QJ (2017) Parametric modeling of microwave components using adjoint neural networks and pole-residue transfer functions with em sensitivity analysis. IEEE Trans Microw Theory Tech 65(6):1955–1975
Zhao Z, Feng F, Zhang W, Zhang J, Jin J, Zhang QJ (2020) Parametric modeling of em behavior of microwave components using combined neural networks and hybrid-based transfer functions. IEEE Access 8:93922–93938
Zhuo Y, Feng F, Zhang J, Zhang QJ (2022) Parametric modeling incorporating joint polynomial-transfer function with neural networks for microwave filters. IEEE Trans Microw Theory Tech 70(11):4652–4665
Feng F, Guo Q, Chen J, Liu W, Zhang W, Zhang J, Na W, Ma K, Zhang QJ (2023) Feature and em sensitivity co-assisted neuro-tf surrogate optimization for microwave filter design. IEEE Trans Microw Theory Tech 71(11):4749–4761
Ye Z, Shao W, Ding X, Wang BZ, Sun S (2023) Knowledge-based neural network for multiphysical field modeling. IEEE Trans Microw Theory Tech 71(5):1967–1976
Zhang W, Feng F, Yan S, Zhao Z, Na W (2020) Multiphysics parametric modeling of microwave components using combined neural networks and transfer function. IEEE Access 8:5383–5392
Wang M, Deng W (2021) Deep face recognition: A survey. Neurocomputing 429:215–244
Ma K, Tang C, Zhang W, Cui B, Ji K, Chen Z, Abraham A (2023) Dc-cnn: Dual-channel convolutional neural networks with attention-pooling for fake news detection. Appl Intell 53(7):8354–8369
Xu X, Wang Q, Guo L, Zhang J, Ding S (2023) Femrnet: Feature-enhanced multi-scale residual network for image denoising. Appl Intell 53(21):26027–26049
Bikram P, Das S, Biswas A (2024) Attentive graph structure learning embedded in deep spatial-temporal graph neural network for traffic forecasting. Appl Intell 1–34
Jin J, Zhang C, Feng F, Na W, Ma J, Zhang QJ (2019) Deep neural network technique for high-dimensional microwave modeling and applications to parameter extraction of microwave filters. IEEE Trans Microw Theory Tech 67(10):4140–4155
Pan G, Wu Y, Yu M, Fu L, Li H (2020) Inverse modeling for filters using a regularized deep neural network approach. IEEE Microw Wirel Compon Lett 30(5):457–460
Ohira M, Takano K, Ma Z (2021) A novel deep-q-network-based fine-tuning approach for planar bandpass filter design. IEEE Microw Wirel ComponLett 31(6):638–641
Zhou Y, Xie J, Ren Q, Zhang HH, Liu QH (2022) Fast multi-physics simulation of microwave filters via deep hybrid neural network. IEEE Trans Antennas Propag 70(7):5165–5178
Ren Y, Deng X, You Z, Zheng X (2024) 1-d multi-channel cnn with transfer functions for inverse electromagnetic behaviors modeling and design optimization of high-dimensional filters. Appl Intell 54(1):503–521
Yu C, Li Q, Feng F, Zhang QJ (2022) Convolutional neural network with adaptive batch-size training technique for high-dimensional inverse modeling of microwave filters. IEEE Microw Wireless Technol Lett 33(2):122–125
Charoosaei F, Noohi M, Sadrossadat SA, Mirvakili A, Na W, Feng F (2022) High-order deep recurrent neural network with hybrid layers for modeling dynamic behavior of nonlinear high-frequency circuits. IEEE Trans Microw Theory Tech 70(12):5340–5358
Faraji A, Noohi M, Sadrossadat SA, Mirvakili A, Na W, Feng F (2022) Batch-normalized deep recurrent neural network for high-speed nonlinear circuit macromodeling. IEEE Trans Microw Theory Tech 70(11):4857–4868
Chaturvedi S, Kumar N, Kumar R (2023) Two feedback pid controllers tuned with teaching–learning-based optimization algorithm for ball and beam system. IETE J Res 1–10
Kumar R, Srivastava S, Mohindru A (2022) Lyapunov stability-dynamic back propagation-based comparative study of different types of functional link neural networks for the identification of nonlinear systems. Soft Comput 24(7):5463–5482
Shobana R, Jaint B, Kumar R (2024) Design of a novel robust recurrent neural network for the identification of complex nonlinear dynamical systems. Soft Comput 28(3):2737–2751
Dass A, Srivastava S, Kumar R (2023) A novel lyapunov-stability-based recurrent-fuzzy system for the identification and adaptive control of nonlinear systems. Appl Soft Comput 137:110161
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 770–778
Zhou Y, Zhang C, Deng L, Fu J, Li H, Xu Z, Zhang J (2024) Resolution-sensitive self-supervised monocular absolute depth estimation. Appl Intell 54(6):4781–4793
Walha R, Aouini A (2024) Scene text image super-resolution using multi-scale convolutional neural network with skip connections. Appl Intell 1–13
Zhou Y, Zhang Y, Xie X, Kung SY (2021) Image super-resolution based on dense convolutional auto-encoder blocks. Neurocomputing 423:98–109
Wang X, Li Z, Huang Y, Jiao Y (2022) Multimodal medical image segmentation using multi-scale context-aware network. Neurocomputing 486:135–146
Ma J, Dang S, Li P, Watkins G, Morris K, Beach M (2022) Transfer learning for the behavior prediction of microwave structures. IEEE Microw Wireless Technol Lett 33(2):126–129
Funding
This work was supported in part by the National Key Technologies Research and Development Program of China under Grant 2023YFB3207800 and in part by the National Natural Science Foundation of China under Grant 61927804.
Author information
Authors and Affiliations
Contributions
Conceptualization: Yimin Ren, Xiaoping Zheng; Methodology: Yimin Ren; Formal analysis and investigation: Yimin Ren; Writing - original draft preparation: Yimin Ren; Writing - review and editing: Xiaojiao Deng, Xiaoping Zheng; Funding acquisition: Xiaoping Zheng; Supervision: Xiaoping Zheng. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors have no relevant financial or non-financial interests to disclose.
Ethics approval and consent to participate
This study does not involve research on humans or animals. Therefore, ethics approval and consent to participate are not applicable.
Consent for publication
All authors consent to the publication of this manuscript.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Ren, Y., Deng, X. & Zheng, X. ISFNN: an enhanced neural network for parametric modeling of passive devices with input skip-connections. Appl Intell 55, 46 (2025). https://doi.org/10.1007/s10489-024-05853-9
Accepted:
Published:
DOI: https://doi.org/10.1007/s10489-024-05853-9