AUTHORS: Jianwei Mi, Xiaoli Fang, Libin Fan

Download as PDF

ABSTRACT: The compact genetic algorithm uses the value of probability variable to represent the population, and each generation produces two chromosomes by the value. This dominant property makes it efficient to be applied in evolvable hardware, and significantly reduces the storage capacity. However, in case of dealing with the complex issues, its execution effects often fail to reach the required demands. Taking the problems above into account, so this paper presents an improved compact genetic algorithm named None-persistent Elitism TCGA (ne-TCGA), combined with the analysis of convergent trend. On the basis of TCGA (Compact Genetic Algorithm with Tendency), it adopts the strategy of non-durable elitist preservation, which both ensured the adequate selection pressure, and maintained the diversity of the population in the evolutionary process. The analysis shows that the ne-TCGA applied in evolvable hardware has better computational efficiency than other random search algorithm.

KEYWORDS: Evolvable hardware, genetic algorithm, ne-TCGA

REFERENCES:

[1] HARTENSTEIN R. Trends in reconfigurable logic and reconfigurable computing. 9th IEEE International Conference on Electronics. DUBROVNIK, CROATIA, 2002: 801-808.

[2] Gordon TGW , Bentley PJ. Towards development in evolvable hardware. NASA/DOD Conference on Evolvable Hardware. ALEXANDRIA, VA, 2002:241- 250.

[3] Fernando PR, Katkoori S, Keymeulen D, Zebulum R, Stoica A. Customizable FPGA IP core implementation of a general-purpose genetic algorithm engine. IEEE Trans Evol Comput. 2010, 14(1):133–149

[4] Yutana Jewajinda, Prabhas Chongstitvatana. A parallel genetic algorithm for adaptive hardware and its application to ECG signal classification.Neural Comput & Applic. 2013, 22:1609–1626

[5] Amaral JFM , Amaral JLM. Towards evolvable analog airtificial neural networks controllers. 6th NASA/DoD Conference on Evolvable Hardware. Jet Propuls Lab, Seattle,2004:46-52.

[6] Kang Li-Shan, He Wei, Chen Yu-Ping. Evolvable hardware realized with function type programmable device. Chinese Journal of Computers, 1999, 22(7): 781- 784

[7] Zhao Shu-Guang, Liu Gui-Xi. Basic basic theory and key technology of evolutionary hardware. System Engineering and Electronic Technique.2002,24(1):70-73.

[8] Marco A. Moreno-Armendáriz, Nareli CruzCortés,Carlos A. Duchanoy,et al, Hardware implementation of the elitist compact Genetic Algorithm using Cellular Automata pseudorandom number generator. Computers and Electrical Engineering,2013(39):1367-1379.

[9] Fernando PR, Katkoori S, Keymeulen D, Zebulum R, Stoica A.Customizable FPGA IP core implementation of a general-purpose genetic algorithm engine. IEEE Trans Evol Comput,2010,14(1):133–149.

[10] Harik G R. The Compact Genetic Algorithm

[J]. IEEE Trans. on Evol. Comput., 1999, 3(4): 287-297.

[11] Cupertino F, Mininno E, Lino E, Naso D. Compact genetic algorithms for the optimization of induction motor cascaded control. Electric machines & drives conference,2007:7–82.

[12] LIU Jieli, YAO Rui. The Implement of Evolvable Hardware on the Design Method of SOPC. Journal of Jiamusi University. 2012,30(2):109-212.