Accurate Modeling of Frequency Selective Surfaces Using Fully-Connected Regression Model With Automated Architecture Determination and Parameter Selection Based on Bayesian Optimization

Çalık, Nurullah; Belen, Mehmet A.; Mahoutı, Peyman; Koziel, Slawomir

doi:10.1109/access.2021.3063523

Cited by 32 publications

(27 citation statements)

References 96 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Four kinds of structural parameters of FsPN contain a) the number of input variables, b) a collocation of the specific subset of input variables, c) the number of membership functions, and d) the type of polynomial. Next, after obtaining the optimized FsPN, the optimization design of both nodes selected in the current layer and their ensuing layer (next layer) leads to the optimized HRmFNN structure [46][47][48][49].…”

Section: Introductionmentioning

confidence: 99%

Hierarchically Reorganized Multi-Layer Fuzzy Neural Networks Architecture Driven With the Aid of Node Selection Strategies and Structural Network Optimization

Wang

Kim

et al. 2022

IEEE Access

View full text Add to dashboard Cite

In this study, a design methodology based on fuzzy sets inference and polynomial neural network(PNN) for hierarchically reorganized self-organizing network architecture is introduced to cope with over-fitting as well as multi-collinearity problems which generally appear in a conventional fuzzy neural network. The design method of the proposed self-organizing network structure provides an efficient solution to construct the hierarchically reorganized multi-layer fuzzy neural networks (HRmFNN) architecture through a synergy of multi-techniques such as L2-norm regularization, probability theory, and multioptimization. The overall network structure is realized with the aid of parallel network structure with newly added inputs as well as effective neuron selection method through the exponential-based roulette selection technique for each layer in HRmFNN, and the least square error estimation (LSE)-based learning method with L2-norm regularization is used for constructing the stabilized network architecture, and their ensuring design methodologies result in alleviating the overfitting phenomenon and also enhancing the generalization ability. For the performance enhancement of HRmFNN directly affected by some parameters such as the number of input variables, collocation of the specific subset of input variables, the number of membership functions per each variable, and the order of polynomial in the consequent parts of the fuzzy rules, multiparticle swarm optimization (MPSO) is exploited for the effectively structural as well as parametric optimization of the proposed network. That is, the multi-optimization helps achieve a compromise between the better generation performance and the alleviated over-fitting leading to the stabilization of the proposed multi-layered self-organizing network structure with the aid of synergistic multi-techniques such as a) L2norm regularization-based LSE learning, b) probability theory for effective neuron selection, and c) novel parallel network structure including newly added inputs and neuron selection method. The performance of the proposed network structure is quantified by comprehensive experiments and comparative analysis. It is also demonstrated through the application to cement compressive strength.

show abstract

Section: Introductionmentioning

confidence: 99%

Hierarchically Reorganized Multi-Layer Fuzzy Neural Networks Architecture Driven With the Aid of Node Selection Strategies and Structural Network Optimization

Wang

Kim

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…With the development of EM technology, absorbers have been applied in civil technical fields, such as energy harvesting, 6 biosensors 7 and microwave chambers, some of which are used to solve the problem of EM radiation pollution 8 . As a research hotspot in many fields, fractal structure has self‐similarity and space filling due to the similarity between local shape and overall shape 9 . It is also widely used in the broadband and miniaturization design of microwave absorber.…”

Section: Introductionmentioning

confidence: 99%

“…8 As a research hotspot in many fields, fractal structure has self-similarity and space filling due to the similarity between local shape and overall shape. 9 It is also widely used in the broadband and miniaturization design of microwave absorber.…”

Section: Introductionmentioning

confidence: 99%

Three‐dimensional printed fractaldielectric‐resonator‐basedbroadband microwave absorber

Ren

Sun

et al. 2022

Int J RF Mic Comp-Aid Eng

View full text Add to dashboard Cite

A broadband microwave absorber based on fractal rectangular dielectric resonator (DR) unit is investigated in this paper. The designed absorber can be fabricated with low-cost three-dimensional printing technology, using the high-loss carbon-carrying acrylonitrile-butadiene-styrene (ABS) polymer. To enhance the operating bandwidth, the fractal-shaped rectangular DR unit is used and multi modes are excited. For verify, a prototype of the absorber with 8 Â 8 unit cells were fabricated and measured. The results indicate that the absorptivity can achieve greater than 90% within a relative bandwidth of 130% from 3.1 to 14.7 GHz, which covers the whole C-band (4-8 GHz) and X-band (8-12 GHz). In addition, the absorption properties of the absorber at different incident angles were studied.

show abstract

“…However, this method is more dependent on multiple feature parameters. Among them, In recent years, machine learning‐assisted modeling is introduced into the design field of microwave and millimeter‐wave devices to reduce the computational pressure brought by full‐wave simulation, such as reinforcement learning, 8 artificial neural network (ANN) 9,10 support vector machine (SVM), 11 Gaussian process regression (GPR) 12 . Among them, ANN is the most widely used, but traditional neural network structures 13,14 have many network layers, slow convergence, and require large amounts of data as training data, which need to be obtained by full‐wave simulation, therefore take longer time.…”

Section: Introductionmentioning

confidence: 99%

Accurate microwave filter design based on particle swarm optimization and one‐dimensional convolution autoencoders

Wang

Zhang

et al. 2021

Int J RF Mic Comp-Aid Eng

View full text Add to dashboard Cite

This paper proposes a one‐dimensional convolutional autoencoders (1D‐CAE) surrogate‐based electromagnetic (EM) optimization technique exploiting particle swarm optimization (PSO) algorithm for microwave filter design. The 1D‐CAE is a flexible neural network structure that can be used as surrogate model to handle forward modeling problems as well as solving inverse problems, such as extracting the coupling matrix from S‐parameters. In this paper, the 1D‐CAE is proposed to handle forward modeling problem. On the one hand, the 1D‐CAE model is adopted to as the surrogate model to the EM simulation solver to speed up the PSO algorithm optimization process. On the other hand, the 1D‐CAE model can update the data set and network parameters online to continuously improve the prediction accuracy of the surrogate model during the PSO algorithm iterations. Compared with other neural network structures, the 1D‐CAE model proposed in this paper requires less data and converges faster in the same test environment. To demonstrate the validity of the proposed method, it was used to design a sixth‐order and an eighth‐order cross‐coupled filter. The design results shown that the proposed method is valid.

show abstract

Accurate Modeling of Frequency Selective Surfaces Using Fully-Connected Regression Model With Automated Architecture Determination and Parameter Selection Based on Bayesian Optimization

Cited by 32 publications

References 96 publications

Hierarchically Reorganized Multi-Layer Fuzzy Neural Networks Architecture Driven With the Aid of Node Selection Strategies and Structural Network Optimization

Hierarchically Reorganized Multi-Layer Fuzzy Neural Networks Architecture Driven With the Aid of Node Selection Strategies and Structural Network Optimization

Three‐dimensional printed fractaldielectric‐resonator‐basedbroadband microwave absorber

Accurate microwave filter design based on particle swarm optimization and one‐dimensional convolution autoencoders

Contact Info

Product

Resources

About