Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming

Research output: Contribution to journalArticle

Standard

Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming. / Walker, James Alfred; Liu, Jerry; Tempesti, Gianluca; Timmis, Jon; Tyrrell, Andy.

In: Journal of Adaptive, Resilient and Autonomic Systems, Vol. 3, No. 4, 2012, p. 32-50.

Research output: Contribution to journalArticle

Harvard

Walker, JA, Liu, J, Tempesti, G, Timmis, J & Tyrrell, A 2012, 'Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming', Journal of Adaptive, Resilient and Autonomic Systems, vol. 3, no. 4, pp. 32-50. https://doi.org/10.4018/jaras.2012100103

APA

Walker, J. A., Liu, J., Tempesti, G., Timmis, J., & Tyrrell, A. (2012). Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming. Journal of Adaptive, Resilient and Autonomic Systems, 3(4), 32-50. https://doi.org/10.4018/jaras.2012100103

Vancouver

Walker JA, Liu J, Tempesti G, Timmis J, Tyrrell A. Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming. Journal of Adaptive, Resilient and Autonomic Systems. 2012;3(4):32-50. https://doi.org/10.4018/jaras.2012100103

Author

Walker, James Alfred ; Liu, Jerry ; Tempesti, Gianluca ; Timmis, Jon ; Tyrrell, Andy. / Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming. In: Journal of Adaptive, Resilient and Autonomic Systems. 2012 ; Vol. 3, No. 4. pp. 32-50.

Bibtex - Download

@article{370a3a23d2fa45188b638edf7b4740ed,
title = "Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming",
abstract = "Transport triggered architectures are used for implement-ing bio-inspired systems due to their simplicity, modularity and fault-tolerance. However, producing ecient, optimised machine code for sucharchitectures is an extremely dicult task, as the computational complex-ity has moved from the hardware-level to the software-level. This paperpresents the application of Cartesian Genetic Programming to the evo-lution of machine code for a simple implementation of a transport trig-gered architecture. The eectiveness of the algorithm is demonstrated byevolving machine code for a 4-bit multiplier with three dierent levels ofparallelism. The results show that 100% successful solutions were foundby CGP and by further optimising the size of the solutions, it is possibleto nd ecient implementations of the 4-bit multiplier. Further analysisof the solutions showed that the use of loops within the CGP function setcould be benecial and was demonstrated by repeating the earlier 4-bitmultiplier experiment with the addition of a loop function. The furtherresults showed that the use of a loop function did not signincatly aectthe performance of the CGP algorithm but a reduction of up to 55%in the solution size was observed, which could have the potential to beclassed as \human competitive{"}.",
author = "Walker, {James Alfred} and Jerry Liu and Gianluca Tempesti and Jon Timmis and Andy Tyrrell",
year = "2012",
doi = "10.4018/jaras.2012100103",
language = "English",
volume = "3",
pages = "32--50",
journal = "Journal of Adaptive, Resilient and Autonomic Systems",
number = "4",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Automatic Machine Code Generation for a Transport Triggered Architecture using Cartesian Genetic Programming

AU - Walker, James Alfred

AU - Liu, Jerry

AU - Tempesti, Gianluca

AU - Timmis, Jon

AU - Tyrrell, Andy

PY - 2012

Y1 - 2012

N2 - Transport triggered architectures are used for implement-ing bio-inspired systems due to their simplicity, modularity and fault-tolerance. However, producing ecient, optimised machine code for sucharchitectures is an extremely dicult task, as the computational complex-ity has moved from the hardware-level to the software-level. This paperpresents the application of Cartesian Genetic Programming to the evo-lution of machine code for a simple implementation of a transport trig-gered architecture. The eectiveness of the algorithm is demonstrated byevolving machine code for a 4-bit multiplier with three dierent levels ofparallelism. The results show that 100% successful solutions were foundby CGP and by further optimising the size of the solutions, it is possibleto nd ecient implementations of the 4-bit multiplier. Further analysisof the solutions showed that the use of loops within the CGP function setcould be benecial and was demonstrated by repeating the earlier 4-bitmultiplier experiment with the addition of a loop function. The furtherresults showed that the use of a loop function did not signincatly aectthe performance of the CGP algorithm but a reduction of up to 55%in the solution size was observed, which could have the potential to beclassed as \human competitive".

AB - Transport triggered architectures are used for implement-ing bio-inspired systems due to their simplicity, modularity and fault-tolerance. However, producing ecient, optimised machine code for sucharchitectures is an extremely dicult task, as the computational complex-ity has moved from the hardware-level to the software-level. This paperpresents the application of Cartesian Genetic Programming to the evo-lution of machine code for a simple implementation of a transport trig-gered architecture. The eectiveness of the algorithm is demonstrated byevolving machine code for a 4-bit multiplier with three dierent levels ofparallelism. The results show that 100% successful solutions were foundby CGP and by further optimising the size of the solutions, it is possibleto nd ecient implementations of the 4-bit multiplier. Further analysisof the solutions showed that the use of loops within the CGP function setcould be benecial and was demonstrated by repeating the earlier 4-bitmultiplier experiment with the addition of a loop function. The furtherresults showed that the use of a loop function did not signincatly aectthe performance of the CGP algorithm but a reduction of up to 55%in the solution size was observed, which could have the potential to beclassed as \human competitive".

U2 - 10.4018/jaras.2012100103

DO - 10.4018/jaras.2012100103

M3 - Article

VL - 3

SP - 32

EP - 50

JO - Journal of Adaptive, Resilient and Autonomic Systems

JF - Journal of Adaptive, Resilient and Autonomic Systems

IS - 4

ER -