27 June 2023 in IEEE Transactions on Electromagnetic Compatibility
In this research article, a frequency selective surface (FSS) for double band stop response for sub 5G (6 GHz) communication, X-band (10 GHz) communication, and band pass response for 7.2 GHz for ultra-wideband applications with optimized dimensions will be predicted by machine learning (ML) and deep learning (DL) techniques. From the equivalent circuit model (ECM), the inductance and capacitance characteristics of various dimensions are calculated, and datasets are prepared. The shielding effectiveness, operating frequency, transmission loss $(S_{21})$ , and return loss $(S_{11})$ are taken as input data, and the structure dimension as output data for the ML and DL algorithms. Two ML and one DL algorithms are tried for getting a mean square error of 0.02–0.09 between the desired and observed results. The size of the FSS element is 0.6 $\lambda _{0}$ × 0.6 $\lambda _{0}$ for the shielding frequency. The thickness of the shielding structure is 1.6 mm, and the dielectric material used is RT Duroid 5880 $(\varepsilon _{r}=2.2)$ . The extreme gradient boosting, adaptive boosting, and deep neural network algorithms are used for multiple input features and multiple output features from the datasets through advanced feature extraction and regression techniques. The results obtained from the static back propagation algorithm are endorsed using EM simulation and analyzed from the fabricated structure.
13 November 2014 in Wireless Personal Communications
1 January 2014 in International Journal of Computer and Communication Engineering
10 January 2025 in IEEE Journal on Multiscale and Multiphysics Computational Techniques
A multi-layer perceptron (MLP) model was applied to electromagnetic shielding to analyze a coupled ring anisotropic frequency selective surface (CRAFSS) using an equivalent circuit model. The shielding structure, based on a single-sided RT 5880 array, features unit elements with dimensions of $0.55\lambda _{0} \times 0.41\lambda _{0}$ at the resonant frequency. Various deep neural network (DNN) configurations with hidden layers were tested to achieve optimal results, reaching a minimal mean square error (MSE) of $1.012 \times 10^{-4}$ . The MLP was trained using input parameters such as S-parameters, resonant frequency, and shielding effectiveness, with the output being the dimensions of the proposed shielding structure. The dataset, built from capacitance and inductance values, was used for testing, training, and validation within the neural network, eventually employing inverse modeling for output prediction. The structure demonstrated stable bandwidth performance despite changes in the incidence angle of transverse magnetic (TM) and transverse electric (TE) polarizations, shifting from $\theta$ = $0^{0}$ to $60^{0}$ . The anisotropic FSS was developed and evaluated, with deep learning results and electromagnetic (EM) simulations playing a key role in the design process.
17 January 2020 in International Journal of RF and Microwave Computer-Aided Engineering