Multi-board Run-time Reconfigurable Implementation of Intrinsic Evolvable Hardware

A multi-board run-time reconfigurable (MRTR) system for evolvable hardware (EHW) is introduced with the aim to implement on hardware the bidirectional incremental evolution (BIE) method. The main features of this digital intrinsic EHW solution rely on the multi-board approach, the variable chromosome length management and the partial configuration of the reconfigurable circuit. These three features provide a high scalability to the solution. The design has been written in VHDL with the concern of not being platform dependant in order to keep a flexibility factor as high as possible. This solution helps tackling the problem of evolving complex task on digital configurable support.




References:
[1] Yao, X., Higuchi, T.: Promises and Challenges of Evolvable Hardware.
IEEE Trans. Systems, Man and Cybernetics, Part C, Vol. 29 (1999) 87 -
97
[2] Goldberg, D. E.: Genetic Algorithm in Search, Optimization and
Machine Learning. Addison-Wesley Publishing Company, Incorporated,
Reading, Massachusetts (1989)
[3] Macias, N. J.: The PIG Paradigm: the Design and Use of a Massively
Parallel Fine Grained Self-Reconfigurable Infinitely Scalable
Architecture. Proc. of the First NASA/DoD Conf. on Evolvable
Hardware, 19-21 (1999) 175 - 180
[4] Ozsvald, I.: Short-Circuiting the Design Process: Evolutionary
Algorithms for Circuit Design Using Reconfigurable Analogue
Hardware. Masters Thesis (1998)
[5] Stoica, A., Keymeulen, D., Vu, D., Zebulum, R., Ferguson, I., Daud, T.,
Arsian, T., Xin, G.: Evolutionary Recovery of Electronic Circuits from
Radiation Induced Faults. CEC2004 conf. on Evolutionary Computation,
Congress on, Vol.: 2, 19-23. Vol.2 (2004) 1786 - 1793
[6] Langeheine, J., Meier, K., Schemmel, J., Trefzer, M.: Intrinsic Evolution
of Digital-to-Analog Converters Using a CMOS FPTA Chip. Proc. of
the Sixth NASA/DoD Conf. on Evolvable Hardware, 24-26 (2004) 18 -
25
[7] Torresen, J.: Evolving Both Hardware Subsystems and the Selection of
Variants of Such Into An Assembled System. In proc. of 16th European
Simulation Multiconference. Darmstadt, Germany (2002) 451 - 457
[8] Baumgarte, V., May, F., N├╝ckel, A., Vorbach, M., Weinhardt, M.:
PACT XPP - A self-Reconfigurable Data Processing Architecture.
Presented at ERSA, Las Vegas, NV, (c) CSREA Press (2001)
[9] Layzell, P.: Reducing Hardware Evolution-s Dependency on FPGAs. In
proc. of MicroNeuro-99, 7th International Conference on
Microelectronics for Neural, Fuzzy and Bio-inspired Systems, IEEE,
Computer Society, CA (1999) 171 - 178
[10] Friedl, S., Sekanina, L.: The First Circuits Evolved in a Physical Virtual
Reconfigurable Device. In: Proc. of the 7th IEEE Workshop on Design
and Diagnostics of Electronic Circuits and Systems, Bratislava, SK,
SAV. ISBN 80-969117-9-(2004) 35 - 42
[11] Glette, K., Torresen, J.: A Flexible On-Chip Evolution System
Implemented on a Xilinx Virtex-II Pro Device. ICES (2005) 66 - 75
[12] Sekanina, L.: Towards Evolvable IP Cores for FPGAs. Proc. of the Fifth
NASA/DoD Conf. on Evolvable Hardware, Los Alamitos, USA, ICSP,
ISBN 0-7695-1977-6 (2003) 145 - 154
[13] Tufte, G., Haddow, P. C.: Towards Development on a Silicon-based
Cellular. Computing Machine, Natural Computing. Vol. 4, Issue 4
(2005) 387 - 416
[14] Xilinx: Virtex 2.5V FPGA Detailed Functional Description. Version
2.8.1 (2002)
[15] Bäck, T., Hoffmeister, F., Schwefel, H. P.: A survey of evolutionary
strategies. In R. Belew and L. Booker, editors, Proc. of the 4th
International Conference on Genetic Algorithms, San Francisco, CA,
1991. Morgan Kaufmann (1991) 2 - 9
[16] Schwefel, H. P.: Numerical Optimization of Computer Models. John
Wiley & Sons, Chichester, UK (1981)
[17] Miller, J.: An Empirical Study of the Efficiency of Learning Boolean
Functions Using a Cartesian Genetic Programming Approach. In Proc.
of the Genetic and Evolutionary Computation Conference. Volume 1,
Orlando, USA (1999) 1135 - 1142
[18] Kalganova, T., Miller, J.: Evolving More Efficient Digital Circuits by
Allowing Circuit Layout Evolution and Multi-objective Fitness. Proc. of
the First NASA/DoD Conf. on Evolvable Hardware, IEEE Computer
Society (1999) 54 - 63
[19] Kalganova, T.: Evolvable Hardware Design for Combinational Logic
Circuits. PhD thesis, School of Computing, Napier University,
Edinburgh, UK (2000)
[20] Xilinx: Virtex-E 1.8V FPGA Detailed Functional Description. Version
2.8 (2006)
[21] Xilinx: Virtex-II Complete Data Sheet. Version 3.4 (2005)
[22] Xilinx: Virtex-4 Data Sheet: DC and Switching Characteristics. Version
2.12 (2006)
[23] Lambert, C., Kalganova, T., Stomeo, E.: FPGA-based Systems for
Evolvable Hardware. ICCS'06 Vienna, Austria, Volume 12, ISBN 975-
00803-1-9 (2006)