Scribd Logo
Upload

Welcome to Scribd!

  • 16 views

    Design of Microstrip Antennas Using Artificial Neural Networks

    Uploaded by

    Divyanshu Bose

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Design of Microstrip Antennas Using Artificial Neural Networks

    Uploaded by

    Divyanshu Bose
    16 views3 pages

    Original Title

    2

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    16 views3 pages

    Design of Microstrip Antennas Using Artificial Neural Networks

    Uploaded by

    Divyanshu Bose

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    International Conference on Computational Intelligence and Multimedia Applications 2007

    Design of Microstrip Antennas Using Artificial Neural Networks


    J.Lakshmi Narayana Dr.K.SriRamaKrishna Dr.L.Pratap Reddy
    Asst.Professor of ECE, Professor & Head of Professor of ECE,
    VR Siddhartha ECE, V R Siddhartha Jawaharlal Nehru.
    Engineering College, Engineering College, Technological
    Vijayawada, A.P, Vijayawada, A.P, University,
    [email protected] [email protected] Hyderabad,A.P,

    Abstract
    A general design procedure is suggested for the micro strip antennas using artificial neural
    networks and this is demonstrated using the rectangular patch geometry. In the design
    procedure, synthesis and analysis are defined as forward side and reverse side of the problem.

    Keywords: micro strip antennas, artificial neural networks, reverse modeling.

    1. Introduction
    In high-performance spacecraft, aircraft, missile and satellite applications, where size,
    weight, cost, performance, ease of installation and aerodynamic profile are constraints, low
    profile antennas may be required. To meet these requirements, microstrip antennas that are low-
    profile, conformable to planar and non-planar surfaces, simple and inexpensive to manufacture
    using modern printed circuit technology, mechanically robust when mounted on rigid surfaces,
    compatible with MMIC designs, and when particular patch shape and mode selected they are
    very versatile in terms of resonant frequency, polarization, pattern and impedance can be used
    [1]. In this work, rectangular microstrip antennas are the ones under consideration .Design
    complexity and non linearity are two major bottle necks in the design of these antennas.ANN
    models have been in the analysis and synthesis of these antennas,which are simple, easy to
    apply and very useful for antenna engineers to predict both patch dimensions and resonant
    frequency[2-5].The analysis problem can be defined as to obtain resonant frequency for a
    given dielectric material and geometric structure (Figure 2). However in this work, the
    corresponding synthesis ANN model is built to obtain patch dimensions of rectangular
    microstrip antennas (W, L) as the function of input variables, which are the height of the
    dielectric substrate (h), dielectric constant of the dielectric material (εr) and the resonant
    frequency ( fr ) (Figure1).
    2.Design problem for the microstrip antenna
    The forward and reverse sides of the problem will be defined for the rectangular patch
    geometry in the following section.

    The Forward Side and Reverse side of the Problem: the Synthesis & the Analysis ANN

    Figure 1. The synthesis ANN model Figure 2. The analysis ANN model.

    0-7695-3050-8/07 $25.00 © 2007 IEEE 332


    DOI 10.1109/ICCIMA.2007.53

    Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on February 05,2021 at 11:21:44 UTC from IEEE Xplore. Restrictions apply.
    3. Building neural networks for the rectangular microstrip antenna and
    Results
    Methods and Simulations

    Both Multilayer Perceptron (MLP) and Radial Basis Function (RBF) networks were
    used in ANN models [7]. MLP models were trained with almost all network learning
    algorithms. Among all these, the ones giving the best results for MLP network has been shown
    in Table 1. MLP network, which has configuration of 3 input neurons, 10 and 5 neurons in
    two hidden layers, and 2 output neurons with learning rate=0.1, goal=0.001, was trained for 500
    epochs. Hyperbolic tangent sigmoid and linear transfer functions were used in MLP training. In
    RBF network, spread value was chosen as 0.01 which gives the best accuracy. Both MLP and
    RBF were trained with 270 samples and tested with 90 samples determined according to the
    definition of the problem; 4 inputs and 1 output were used for the analysis ANN and 3 inputs
    and 2 outputs for the synthesis ANN.
    In Table 1 and Table 2, the accuracy values of synthesis ANN and analysis ANN for
    four networks giving the best results have been given. As can be seen from Table 1 and Table
    2, in synthesis RBF network and in analysis Levenberg-Marquardt network were the one giving
    the best approximation to the target values whose structure is defined in the following
    subsection. The train and test data of the synthesis and analysis ANN were obtained from both
    experimental results given in previous works [4] and a computer program using formulae [6].
    The data are in a matrix form consisting inputs and target values and arranged according to the
    definitions of the problems. The results of the synthesis and analysis ANN and comparison with
    the targets are given in Table 3 and 4, respectively.
    % %
    Accuracy Accuracy
    RBF 99.991 MLP3 99.999
    MLP1 99.974 RBF 99.68
    MLP2 99.162 MLP2 98.151
    MLP3 82.337 MLP1 97.243

    Table 1: Accuracies of the synthesis Table 2: Accuracies of the Analysis


    ANN for Four Networks giving the ANN for Four Networks giving the
    best Results best Results

    RBF: Radial basis function network

    MLP1: Multilayer perceptron network using Conjugate Gradient back propagation as


    learning Algorithm

    MLP2: Multilayer perceptron network using Momentum back propagation as learning


    Algorithm

    MLP3: Multilayer perceptron network using Levenberg-Marquardt optimization as


    learning Algorithm

    333

    Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on February 05,2021 at 11:21:44 UTC from IEEE Xplore. Restrictions apply.
    Table 3. Results of the synthesis ANN and comparison with the targets

    fr (GHz) h (mm) εr W_Target W_RBF W_MLP1 W_MLP2 W_MLP3


    (mm) (mm) (mm) (mm) (mm)
    4.5 15 3.5 22.22 22.54 24.43 22.00 21.82
    3.5 25 8.5 19.66 19.96 19.09 23.19 -3.46
    1.5 25 8.5 45.88 47.52 45.29 47.39 43.57
    3.5 15 6 22.91 23.16 23.96 24.09 25.61
    3.5 22.5 6 22.91 23.29 23.49 24.31 22.34

    Table 4. Results of the analysis ANN and comparison with the targets.

    h (mm) εr W L fr fr fr fr fr
    (mm) (mm) Target MLP3 RBF MLP2 MLP1
    (GHz) (GHz) (GHz) (GHz) (GHz)
    20 3.5 22.22 5.95 4.5 4.50 4.63 4.51 4.58
    25 8.5 12.51 -1.77 5.5 5.50 5.32 4.68 4.61
    25 3.5 28.57 8.04 3.5 3.50 3.44 3.87 3.90
    25 3.5 40 16.97 2.5 2.50 2.52 2.77 2.86
    17.5 8.5 15.29 3.12 4.5 4.50 4.74 4.39 4.34

    4. Conclusion
    In this paper, the neural network is employed as a tool in design of the microstrip
    antennas. In this design procedure, synthesis is defined as the forward side and then analysis as
    the reverse side of the problem. Therefore, one can obtain the geometric dimensions with high
    accuracy, which are the length and the width of the patch in our geometry, at the output of the
    synthesis network by inputting resonant frequency, height and dielectric constants of the chosen
    substrate. Furthermore, in our work, the synthesis can also be applied into anisotropic dielectric
    substrate. In this work the analysis is considered as a final stage of the design procedure,
    therefore the parameters of the analysis ANN network are determined by the data obtained
    reversing the input-output data of the synthesis network. Thus, resonant frequency resulted
    from the synthesized antenna geometry is examined against the target in the analysis ANN
    network. Finally, in this work, a general design procedure for the microstrip antennas is
    suggested using artificial neural networks and this is demonstrated using the rectangular patch
    geometry.
    5. References
    [1] C. A. Balanis, Antenna Theory, John Wiley & Sons, Inc., 1997.
    [2] S. Sagiroglu, K. Güney, Calculation of resonant frequency for an equilateral triangular microstrip antenna using
    artificial neural Networks, Microwave Opt. Technology Lett., Vol. 14, pp. 89-93, 1997.
    [3] S. Sagiroglu, K. Güney and M. Erler, Resonant frequency calculation for circular microstrip antennas using
    artificial neural networks, International Journal of RF and Microwave Computer-Aided Engineering,Vol. 8, No.
    3, pp. 270-277, 1998.
    [4] D. Karaboga, K. Güney, S. Sagiroglu and M. Erler, Neural computation of resonant frequency of electrically thin
    and thick rectangular microstrip antennas, Microwaves, Antennas and Propagation, IEEE Proceedings-Vol. 146,
    No. 2, pp. 155 – 159, April 1999.
    [5] K. Güney, S. Sagiroglu and M. Erler, Generalized neural method to determine resonant frequencies of various
    microstrip antennas, International Journal of RF and Microwave Computer-Aided Engineering, Vol. 12, No. 1,
    pp. 131-139, January 2002.
    [6] I. J. Bahl and P. Bhartia, Microstrip Antennas, Artech House,Dedham , MA, 1980.
    [7] Q. J. Zhang and K. C. Gupta, Neural Networks for RF and Microwave Design, Artech House Publishers, 2000.

    334

    Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on February 05,2021 at 11:21:44 UTC from IEEE Xplore. Restrictions apply.

    You might also like