Neurospectral computation for the resonant characteristics of microstrip patch antenna printed on uniaxially anisotropic substrates

Abstract

Modeling and design of rectangular microstrip patch printed on isotropic or anisotropic substrate are accomplished in this paper. The use of spectral domain method in conjunction with artificial neural networks (ANNs) to compute the resonant characteristics of rectangular microstrip patch printed on isotropic or anisotropic substrates. The moment method implemented in the spectral domain offers good accurateness, but its computational cost is high owing to the evaluation of the slowly decaying integrals and the iterative nature of the solution process. The paper introduces the electromagnetic knowledge combined with ANN in the analysis of rectangular microstrip antenna on uniaxially anisotropic substrate to reduce the complexity of the spectral domain method and to minimize the CPU time necessary to obtain the numerical results. The numerical comparison between neurospectral and conventional moment methods shows significant improvements in time convergence and computational cost. Hence, the use of neurospectral approach presented here as a promising fast technique in the design of microstrip antennas.

Analysis of C-shaped Compact Microstrip Antennas Using Deep Neural Networks Optimized by Manta Ray Foraging Optimization with Lévy-Flight Mechanism

“…As seen in Figure 5 and Figure 6, while the harmony is high at low frequencies, the discrepancy between the estimated data and the simulation data increases as the frequency increases. As mentioned in the literature review, most of the studies related to CCMAs have focused on optimizing the antenna designs for specific applications and obtaining mathematical models that usually provide the resonant frequency [3,5,6,[17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34]. Although this study is not directly related to the estimation of resonant frequencies of CCMAs, the resonant frequencies in dominant mode (TM010) were indirectly estimated by the interpretation of the S11 curves, and the comparison between the results obtained in this study and the results in the literature [7,23] is shown in Table 5.…”

“…In recent years, researchers have vigorously challenged the accurate calculation of the resonant frequency of CMAs. While most studies focus on designing and optimizing CMA structures [3,[17][18][19][20] and obtaining mathematical models for determining the resonant frequencies of the antennas [5,7,[21][22][23][24], there are also studies on computing the resonant frequencies or other characteristic parameters using neural network models [25][26][27][28][29][30][31][32][33][34]. In most recent studies, resonant frequency estimation was made in two ways: the closed-form expression approach and the neural model approach.…”

“…The authors have used isotropic substrate and uniaxially anisotropic substrate in MA design [25]. In another study, a neural model for the synthesis and analysis of rectangular MAs using uniaxially anisotropic substrates was developed by Barkat et al [26]. The patch dimensions were used for synthesis, and the resonant frequency and bandwidth parameters were used for analysis [26].…”

“…In another study, a neural model for the synthesis and analysis of rectangular MAs using uniaxially anisotropic substrates was developed by Barkat et al [26]. The patch dimensions were used for synthesis, and the resonant frequency and bandwidth parameters were used for analysis [26]. In [30], Kaur and Sivia have proposed a miniaturized hybrid fractal antenna based on Giuseppe Peano and Cantor set fractals for biomedical systems.…”