Inverse Modeling for Filters Using a Regularized Deep Neural Network Approach

Extracting coupling matrix from given ${S}$ -parameters can be viewed as an inverse problem for microwave filters, which is of importance for filter design and tuning. In this letter, a regularized deep belief network (R-DBN) is proposed to handle this inverse modeling problem. The training of an R-DBN consists of two steps. First, in unsupervised training, to accommodate the characteristics of input data, this model is constructed with a series of traditional restricted Boltzmann machines (RBMs), which are equipped with a continuous version of transfer function for continuous data processing. In addition, this training can provide suitable weights and bias for the following step. Second, in supervised training, Bayesian regularization is employed to increase modeling ability and prevent overfitting. Two experiments with different simulation environments are illustrated, and the calibration results show high accuracy and robustness in a more intelligent way using this method.

[1]  Weicong Na,et al.  Deep Neural Network Technique for High-Dimensional Microwave Modeling and Applications to Parameter Extraction of Microwave Filters , 2019, IEEE Transactions on Microwave Theory and Techniques.

[2]  Michael Unser,et al.  Deep Convolutional Neural Network for Inverse Problems in Imaging , 2016, IEEE Transactions on Image Processing.

[3]  Junfei Qiao,et al.  A regularization-reinforced DBN for digital recognition , 2017, Natural Computing.

[4]  S. Koziel,et al.  Space Mapping With Multiple Coarse Models for Optimization of Microwave Components , 2008, IEEE Microwave and Wireless Components Letters.

[5]  G. Macchiarella,et al.  A formulation of the Cauchy method suitable for the synthesis of lossless circuit models of microwave filters from lossy measurements , 2006, IEEE Microwave and Wireless Components Letters.

[6]  Chi-Yang Chang,et al.  A Vector-Fitting Formulation for Parameter Extraction of Lossy Microwave Filters , 2007, IEEE Microwave and Wireless Components Letters.

[7]  M. Salazar-Palma,et al.  Interpolation and Extrapolation of S-Parameter Data of a Microwave Filter in the Frequency Domain Using the Cauchy Method , 2018, 2018 IEEE/MTT-S International Microwave Symposium - IMS.

[8]  Qi-Jun Zhang,et al.  Neural Network Inverse Modeling and Applications to Microwave Filter Design , 2008, IEEE Transactions on Microwave Theory and Techniques.

[9]  L. Accatino.  Computer-Aided Tuning of Microwave Filters , 1986, 1986 IEEE MTT-S International Microwave Symposium Digest.

[10]  Qi-Jun Zhang,et al.  High-Dimensional Neural-Network Technique and Applications to Microwave Filter Modeling , 2010, IEEE Transactions on Microwave Theory and Techniques.

[11]  Yongsheng Ou,et al.  Reinforcement learning approach to learning human experience in tuning cavity filters , 2015, 2015 IEEE International Conference on Robotics and Biomimetics (ROBIO).

[12]  Wei Hong,et al.  Efficient Design of Compact Contiguous-Channel SIW Multiplexers Using the Space-Mapping Method , 2015, IEEE Transactions on Microwave Theory and Techniques.

[13]  Vicente E. Boria,et al.  On Space Mapping Techniques for Microwave Filter Tuning , 2019, IEEE Transactions on Microwave Theory and Techniques.

[14]  Peng Li,et al.  Hybrid modeling of microwave devices using multi‐kernel support vector regression with prior knowledge , 2015 .

[15]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[16]  S. Amari,et al.  Space-mapping optimization of planar coupled-resonator microwave filters , 2006, IEEE Transactions on Microwave Theory and Techniques.

[17]  Weicong Na,et al.  Multivalued Neural Network Inverse Modeling and Applications to Microwave Filters , 2018, IEEE Transactions on Microwave Theory and Techniques.