Modelling and Optimization of Metamaterial Antennas using Kernel Machine Approach
Keywords:
Metamaterial, Kernel machine, Antenna, Optimization, ElectromagneticAbstract
This research work is concerned with the application of Gaussian process regression kernel machine learning approach in modelling a metamaterial antenna. MATLAB Graphical user interface (GUI) was used to randomly select two frequency points within the range 2.5GHz ? f ? 3GHz and ten points in the range 3GHz ? f ?3.6GHz respectively. The training and test inputs generated for the model were standardized along with the frequency of operation. The rand function in MATLAB was used to generate ten sets of values between 0 and 1. The lowest negative log likelihood value gotten was -1.9762. This value is produced by the gp model with an initial hyper parameter value set at 0.1386. The model likelihood value is -13.8155. The model is capable of predicting optimal antenna responses for any new set of input data associated with the modelled metamaterial antennas.
References
2. Silvester PP, Ferrari RL, Finite Elements for Electrical Engineers, Cambridge University Press. New York, 1983.
3. Jin J, The Finite-Element Method in Electromagnetics, Wiley, New York, 1993.
4. A. Taflove, SC Hagness (Eds.). Computational Electrodynamics: The Finite-Difference Time-Domain Method, Artech House, Boston, 2005.
5. GJ Burke, AJ Poggio, “Numerical Electromagnetics Code (NEC)-method of moments,” Technical Document 116, Naval Ocean Systems Centre, San Diego, CA, September 6, 1982.
6. G Eleftheriades, KG Balmain, Negative-Refraction Metamaterials: Fundamental Principles and Applications, New York: Wiley, 2006.
7. D Sievenpiper, High-impedance electromagnetic surfaces, Ph.D. dissertation, Department of Electrical Engineering, University of California, Los Angeles, 1999.
8. CA Balanis (Ed.). Modern Antenna Handbook, Wiley, New York, 2008.
9. J.P De Villiers, J.P Jacobs. “Gaussian process modelling of CPW-fed slot antennas,” Progress in Electromagnetics Research 2009; 98: 233-249.
10. B. Liu, and C. Ji, “Automated Metamaterial Design with Computer Model Emulation and Bayesian Optimization,” Article in Applied Mechanics and Materials Journal, August 2014.
11. B. Liu, C. Ji. “Bayesian Nonparametric Modelling for Rapid Design of Metamaterial Microstructures,” International Journal of Antennas and Propagation, 2014.
12. VN Vapnik, The Nature of Statistical Learning Theory. New York: Springer, 1995.
13. CM Bishop. Neural Networks for Pattern Recognition. Oxford: Oxford University Press, 1995.
14. RO Duda, PE Hart, DG Stork. Pattern Classification, 2nd ed. New York: Wiley 2001.
15. C. Gao, Y. Wang. “Analysis of EBG Structures Implemented on CPW Components by Using EMANN Models,” Antennas and Propagation Society International Symposium, 2002. IEEE. 16-21 June 2002.
16. E.Q. Fernandes, H. da Fonseca Silva, A.G. D’Assuncao, and M.A. Barbosa de Melo, “Neuro Modelling of Microwave waveguides with EBG Structures,” Proceedings SBMO/IEEE MTT-S IMOC 2003.
17. M.M. Raman, C. Christodoulou. Support Vector Machines for Antenna Array Processing and Electromagnetics. Synthesis Lectures on Computational Electromagnetics: Morgan and Claypool Publishers. 2006.
18. G. Angiulli, M. Cacciola, M. Versaci. “Microwave devices and Antenna Modelling by Support Vector Regression Machines,” Electromagnetic Field Computation, 2006 12th Biennial IEEE Conference 30 April-3 May 2006.
