Artificial Neuraln Network based Design of Modified Shaped Patch Antenna

Kaur, Rajvinder; Narula, Ashwani Kumar

doi:10.5815/ijisa.2017.04.04

Cited by 3 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kumar et al studied on frequency reconfigurable slot antenna by back propagated ANN algorithm in 2‐5 GHz frequency band 15 . Moreover, modified shaped patch antenna design parameters for desired frequency and gain were investigated by ANN approach in literature 16 . With this perspective, Pattnaik et al investigated the design of a circular fractal antenna for 2.45 GHz ISM band applications; they used a hybrid method, which includes an ANN model in the optimization processes 17 .…”

Section: Introductionmentioning

confidence: 99%

“…15 Moreover, modified shaped patch antenna design parameters for desired frequency and gain were investigated by ANN approach in literature. 16 With this perspective, Pattnaik et al investigated the design of a circular fractal antenna for 2.45 GHz ISM band applications; they used a hybrid method, which includes an ANN model in the optimization processes. 17 In another literature study, Bayesian regularization based ANN model was developed to design a flexible patch antenna for wideband applications.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Ground plane design configuration estimation of 4.9 GHz reconfigurable monopole antenna for desired radiation features using artificial neural network

Alkurt

Ozdemir

Akgöl

et al. 2021

Int J RF Mic Comp-Aid Eng

View full text Add to dashboard Cite

This paper presents a system based on artificial neural network (ANN) that predicts ground plane design for desired radiation properties of a monopole antenna with operation band of 4.8 to 5 GHz. The operating frequency can be adapted to any other frequency regimes. Initially, a 180 Â 180 mm2 ground plane, which is composed of a copper layer, is designed and integrated to a radiative pole that creates monopole antenna configuration. The ground plane is divided into 18 rows and 18 columns as 18 Â 18 matrix so that each unit cell has a square shape having 10 mm side length. Moreover, 152 different ground plane configurations are created by using logic 1 s and 0 s. Multi-layered feed forward ANN is used along with Scale Conjugate Gradient learning algorithm to design ground plane of the monopole antenna. Simulated 152 random ground plane arrays and obtained radiation patterns are used to train ANN for the ground plane design. If a user wants to manipulate radiation, artificial neural network gives the optimum ground plane design for the desired radiation direction and gain with 91.03% accuracy. Finally, one test antenna is fabricated and experimentally measured to support the results of the simulated one. The proposed ANN model approach can be easily used for antenna applications in the antenna industry. K E Y W O R D Santenna design, artificial neural network, monopole antenna, optimized ground plane | INTRODUCTIONRecently, artificial neural networks (ANN) have become very interesting in all areas of engineering applications [1][2][3][4] especially in antenna engineering. [5][6][7][8][9][10] Applications towards to the design parameter estimations of circular patch antennas with the help of ANN algorithms were presented by researchers, 5,6 and also modeling and designs of slotted fractal antenna and Wi-Fi antenna were investigated by using ANNs. 7,8 Moreover, Neebha

show abstract

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

Ground plane design configuration estimation of 4.9 GHz reconfigurable monopole antenna for desired radiation features using artificial neural network

Alkurt

Ozdemir

Akgöl

et al. 2021

Int J RF Mic Comp-Aid Eng

View full text Add to dashboard Cite

show abstract

“…Artificial neural networks can be used as the universal approximators [13] for solving various mathematical problems as numerical analysis tool. For example, a modified shaped patch antenna is analyzed using a feed forward back propagation artificial neural network [14]. Power amplifiers were analyzed using a combination of several neural networks [15], each network was used for prediction of certain parameter of an amplifier.…”

Section: Introductionmentioning

confidence: 99%

Delay Systems Synthesis using Multi-Layer Perceptron Network

et al. 2018

View full text Add to dashboard Cite

The aim of this paper is to accelerate development and investigation of the delay systems. The computational time for investigation of particular design of delay system may take from several minutes up to several days. To achieve the required constructional parameters of the system, the iterative calculations usually should be repeated many times. In this paper, an artificial neural network is proposed to be used as the universal approximator for solving mathematical problems of delay system investigation instead of usual analytical and numerical techniques. The application of a multi-layer perceptron is proposed for approximation of solution space with discrete estimates, which were initially received by application of numerical techniques. Different structures of the multi-layer perceptron were tested for approximation. The difference between delay systems synthesis, which was estimated using numerical techniques and trained multi-layer perceptron did not exceed 5% for any of the six design parameter values. The execution time for estimating single delay system was reduced from 240 s to 20 ms. Such fast estimation of design parameters enables performing delay system analysis and design in real time, preserving time for structure visualization in 3D or virtual reality environment.

show abstract

“…The backpropagation (BP) learning algorithm has gained popularly due to its simplicity, but the slower convergence and getting stuck into local minima are the major weaknesses for degrading its performance. Therefore, some of other proposals have been given, like modified error function [20,21], addition of variable learning rate [22,23], addition of momentum [24,25], delta-bar-delta algorithm [26,27], Levenberg Marquardt (LM) algorithm [51], GA-MLP hybrid algorithm [56], and quick prop [28], to overcome the above issues, but they have not accelerated the convergence to a significant amount. The fast convergence with efficacious performance along with less complexity of neural network is the important matter for the variety of applications.…”

Section: Introductionmentioning

confidence: 99%

On the Root-Power Mean Aggregation Based Neuron in Quaternionic Domain

Kumar¹,

Tripathi²

2018

IJISA

View full text Add to dashboard Cite

This paper illustrates the new structure of artificial neuron based on root-power means (RPM) for quaternionic-valued signals and also presented an efficient learning process of neural networks with quaternionic-valued root-power means neurons (ℍ-RPMN). The main aim of this neuron is to present the potential capability of a nonlinear aggregation operation on the quaternionic-valued signals in neuron cell. A wide spectrum of aggregation ability of RPM in between minima and maxima has a beautiful property of changing its degree of compensation in the natural way which emulates the various existing neuron models as its special cases. Further, the quaternionic resilient propagation algorithm (ℍ-RPROP) with error-dependent weight backtracking step significantly accelerates the training speed and exhibits better approximation accuracy. The wide spectrums of benchmark problems are considered to evaluate the performance of proposed quaternionic rootpower mean neuron with ℍ-RPROP learning algorithm.

show abstract

Artificial Neuraln Network based Design of Modified Shaped Patch Antenna

Cited by 3 publications

References 10 publications

Ground plane design configuration estimation of 4.9 GHz reconfigurable monopole antenna for desired radiation features using artificial neural network

Ground plane design configuration estimation of 4.9 GHz reconfigurable monopole antenna for desired radiation features using artificial neural network

Delay Systems Synthesis using Multi-Layer Perceptron Network

On the Root-Power Mean Aggregation Based Neuron in Quaternionic Domain

Contact Info

Product

Resources

About