An in-system routing strategy for evolvable hardware programmable platforms

On of the major limiting factors for the development of hardware platforms able to support evolvable hardware principles is the lack of simple and compact in-system dynamic routing strategies. In this paper we shall present a programmable hardware architecture whose internal organization permits to perform dynamic routing processes. The architecture is based on a regular bi-dimensional array of functional cells. A hierarchical layered organization has been provided for these cells. Specific routing resources have been included in one of these layers, so that they permit to construct in an incremental way routing paths among the functional cells. The dynamic routing strategy is based on a replication process that is able to connect a source cell with various target cells. One of the major advantages of the proposed routing strategy lies in the fact that its complexity grows only linearly with the array size. Furthermore it is scalable, accommodating without performance degradation to any array size. Behavioral hardware descriptions haven been created for the functional cells that constitute the array. As the simulation and synthesis results will show, the proposed routing strategy will permit the implementation of actual evolvable hardware principles.

[1]  Marco Tomassini,et al.  A phylogenetic, ontogenetic, and epigenetic view of bio-inspired hardware systems , 1997, IEEE Trans. Evol. Comput..

[2]  Moshe Sipper,et al.  Evolution of Parallel Cellular Machines , 1997, Lecture Notes in Computer Science.

[3]  Tom Kean,et al.  XC6200 FastmapTM Processor Interface , 1995, FPL.

[4]  Jordi Madrenas,et al.  Realization of self-repairing and evolvable hardware structures by means of implicit self-configuration , 1999, Proceedings of the First NASA/DoD Workshop on Evolvable Hardware.

[5]  Kiyoshi Oguri,et al.  Plastic cell architecture: towards reconfigurable computing for general-purpose , 1998, Proceedings. IEEE Symposium on FPGAs for Custom Computing Machines (Cat. No.98TB100251).

[6]  Marco Tomassini,et al.  Phylogeny, Ontogeny, and Epigenesis: Three Sources of Biological Inspiration for Softening Hardware , 1996, ICES.

[7]  P. Chow,et al.  The design of an SRAM-based field-programmable gate array. I. Architecture , 1999, IEEE Trans. Very Large Scale Integr. Syst..

[8]  Jonathan Rose,et al.  A detailed router for field-programmable gate arrays , 1990, 1990 IEEE International Conference on Computer-Aided Design. Digest of Technical Papers.

[9]  Jonathan Rose,et al.  The Design of an SRAM-Based Field-Programmable Gate Array — Part I : Architecture , 1999 .

[10]  Nicholas J. Macias,et al.  The PIG paradigm: the design and use of a massively parallel fine grained self-reconfigurable infinitely scalable architecture , 1999, Proceedings of the First NASA/DoD Workshop on Evolvable Hardware.

[11]  Luca Maria Gambardella,et al.  Ant Algorithms for Discrete Optimization , 1999, Artificial Life.