Impacts of Genetic Algorithm Parameters on the Solution Performance for the Uniform Circular Antenna Array Pattern Synthesis Problem

 

F. Yaman¹ , A. E. Yilmaz*²

 

^1,2 Electronics Engineering Department, Ankara University, Tandogan, Ankara, Turkey, 06100 *E–mail: aeyilmaz@eng.ankara.edu.tr fyaman@eng.ankara.edu.tr

 

ABSTRACT

In this paper, the uniform circular antenna array pattern synthesis problem is solved by means of the real coded genetic algorithm (GA). At the same time, the impacts of the mutation rate and the crossover position on the GA performance are also investigated. For this purpose, a circular antenna array with uniformly spaced isotropic elements having identical excitation amplitudes is used as a model. Unlike the conventional GA (with fixed mutation rate and random crossover positions), typical GA implementations with variable mutation rate and restricted crossover position are considered for performance improvement. In conclusion, for the specific problem, decreasing mutation rate with negative derivative is observed to be outperforming the implementations with different mutation rate behaviors. Moreover, regarding the crossover technique, it is observed that imposing some restrictions on the crossover positions (rather than fully random position selection) yields better solutions.

Keywords: Circular antenna array, pattern synthesis, genetic algorithm, mutation rate, crossover point.

 

RESUMEN

En este trabajo, se le da solución a un problema de síntesis de patrones de arreglo de antenas circular uniforme por medio del algoritmo genético con codificación real (GA). Se investigan, al mismo tiempo, los impactos del índice de mutación y la posición de cruce sobre el desempeño del GA. Con tal propósito, se utiliza un arreglo de antenas circular con elementos isotrópicos espaciados uniformemente con amplitudes de excitación idénticas. A diferencia del GA convencional (con índice de mutación y posiciones de cruce aleatorias), se consideran implementaciones de GA típicas con índice de mutación variable y posición de cruce restringida para la mejora del desempeño. En conclusión, para el problema en cuestión, se observa que un índice de mutación descendiente con derivativa negativa supera las implementaciones con comportamientos de índice de mutación diferentes. Además, con relación a la técnica de cruce, se observa que imponer algunas restricciones sobre las posiciones de cruce (en lugar de la selección de posición completamente aleatoria) arroja mejores soluciones.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

References

[I] Holland, J., Adaptation in natural and artificial systems, University of Michigan Press, Ann Arbor, 1975.         [ Links ]

[2] Goldberg, D.E., Genetic algorithms in search optimization and machine learning, Addison–Wesley Longman, 1989.         [ Links ]

[3] Michalewicz, Z., Genetic algorithms + data structures = Evolution programs, Springer–Verlag, Berlin, 1992.         [ Links ]

[4] Whitley, L.D., Foundations of genetic algorithms–2, Morgan Kaufmann Publishers Inc., California, 1993.         [ Links ]

[5] Balanis, C.A., Antenna theory analysis and design, John Wiley & Inderscience Publications, New Jersey, 2005.         [ Links ]

[6] Visser, H.J., Array and phased array antenna basics, John Wiley & Sons Publications, 2005.         [ Links ]

[7] Panduro, M.A., Mendez, A.L., Dominguez, R., Romero, G, Design of non–uniform circular antenna arrays for side lobe reduction using the method of genetic algorithms, AEÜ – International Journal of Electronics and Communications, vol. 60, pp. 713–717, 2006.         [ Links ]

[8] Shihab, M., Najjar, Y., Dib, N., Khodier, M., Design of non uniform circular antenna arrays using particle swarm optimization, Journal of Electrical Engineering, vol. 59, no. 4, pp. 216–220, 2008.         [ Links ]

[9] Xu, Z., Li, H., Liu, Q.–Z., Pattern synthesis of conformal antenna array by the hybrid genetic algorithm, Progress in Electromagnetics Research (PIER), vol. 79, pp. 75–90, 2008.         [ Links ]

[10] Srinivas, M. and Patnaik, L.M., Adaptive probabilities of crossover and mutation in genetic algorithms, IEEE Transactions on Systems, Man and Cybernetics, vol. 24, no. 4, pp. 656–667, April 1994.         [ Links ]

[11] Stanhope, S.A. and Daida, J.M., An individually variable mutation rate strategy for genetic algorithms, Proceedings of the 6th International Conference on Evolutionary Programming VI (Lecture Notes in Computer Science), vol. 1213, pp. 235–246, 1997.         [ Links ]

[12] Pham, D.T. and Karaboga, D., Genetic algorithms with variable mutation rates: Application to fuzzy logic controller design, Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, vol. 211, no. 2, pp.157–167, 1997.         [ Links ]

[13] Marsili Libelli, S., Alba, P., Adaptive mutation in genetic algorithms, Soft Computing, vol. 4, pp. 76–80, 2000.         [ Links ]

[14] Thierens, D., Adaptive mutation rate control schemes in genetic algorithms, Proceedings of IEEE Congress on Evolutionary Computation (CEC '02), pp. 980–985, 2002.         [ Links ]

[15] Uyar, S., Eryigit, G., Sariel, S., An adaptive mutation scheme in genetic algorithms for fastening the convergence to the optimum, Proceedings of the 3rd Asia Pacific International Symposium on Information Technology, pp. 461–465, 2004.         [ Links ]

[16] Soyturk, M., Evaluation of canonical algorithms and adaptive algorithms for two known problems, Journal of Naval Science and Engineering, vol. 5, no. 3, pp. 26–37, 2009.         [ Links ]

[17] Lee, S. and Rowlands, H., Introducing the swingometer crossover and mutation operators for floating–point encoded genetic algorithms, Proceedings of the 19th European Conference on Modelling and Simulation (ECMS 2005), , 2005.         [ Links ]

[18] Jabeen, S.D. and Bhunia, A.K., Real–coded genetic algorithm with variable rates of cross–over and mutation: a basis of global optimization for multi–modal functions via interval technique, International Journal of Computer Mathematics, vol. 83, no. 12, pp. 853–866, 2006.         [ Links ]

[19] Lazarova, M., Borovska, P., Mabgar, S., The impact of the mutation strategy on the quality of solution of parallel genetic algorithms, Proceedings of the 9th WSEAS International Conference on Evolutionary Computing, pp. 63–68, 2008.         [ Links ]

[20] Hinterding, R., Gielewski, H., Peachey, T.C., The nature of mutation in genetic algorithms, Proceedings of the 6th International Conference on Genetic Algorithms, pp 65–72, 1998.         [ Links ]

[21] Piszcz, A. and Soule, T., Genetic programming: Analysis of optimal mutation rates in a problem with varying difficulty, International Journal of Innovative Computing and Applications, vol. 1, no. 2, pp. 108–120, 2006.         [ Links ]

[22] Vrajitoru, D., Crossover improvement for the genetic algorithm in information retrieval, Information Processing & Management, vol. 34, no. 4, pp. 405–415, 1998.         [ Links ]

[23] Shuhai, F., Two problems in genetic algorithm with chromosome differentiation, Proceedings of the 2nd International Symposium on Knowledge Acquisition and Modeling (KAM '09), pp. 248–250, 2009.         [ Links ]

[24] Akbulut, M.B., Yilmaz, A.E., A modified genetic algorithm for the generalized assignment problems, Istanbul University – Journal of Electrical and Electronics Engineering (IU–JEEE), vol. 9, no.2, pp. 951–958, 2009.         [ Links ]