Abstract
The design process of an 8-bit 32-processing-units serial pipeline VLSI modular processor performing the parallel execution of a kernel classifier recall phase is described. Starting from VHDL high-level modelling, after verifying the correct circuit behaviour by means of digital simulation, the circuit has been synthesized and mapped on a 0.7 micron CMOS technology. Both area and delay optimized syntheses are performed for each processor cell, selecting in each case the best solution. Taking advantage of the constant data flow of the pipeline architecture, a dynamic realization of the memory elements using tri-state standard cells is proposed. This reduces both the circuit area and delays, without losing the convenience of an automatic standard cell placement and routing. From synthesis results, a working frequency of about 300 MHz is expected. A parallel-serial interface reduces external clock frequency requirements, and thus matching the external frequency limitations with fast on-chip processing.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Lippmann R., “An Introduction to Computing with Neural Nets”, IEEE Acoustics, Speech and Signal Processing Magazine, 4(2):4–22, April 1987.
D.R. Hush, B.G. Horne, “Progress in Supervised Neural Networks, What's New Since Lippmann?”, IEEE Signal Processing Magazine, pp. 8–39, January 1993.
Comon P., “Classification Supervisée par Réseaux Multicouches”, Traitement du Signal, Vol. 8, No. 6, pp. 387–407, December 1991.
Avilés C. et al., Enhanced Learning for Evolutive Neural Architectures, Project Results and Industrial Openings. Ed. C. Jutten et al., D facto, 1995.
R.O. Duda, P.E. Hart, Pattern Classification and Scene Analysis, John Wiley & Sons, 1972.
Madrenas J., Moreno J.M., Cabestany J., Seco J., “A Bit-Serial VLSI Processor for Kernel-Based Classifiers”, Proc. of the IEEE Workshop on Neural Networks for Signal Processing NNSP'96, IEEE, Kyoto, 1996.
Voz J.L. et al., “Suboptimal Bayesian classification by vector quantization with small clusters”, Proc. of the European Symp. on Artificial Neural Networks ESANN'95, Editor: M. Verleysen, D facto, pp.153–160, 1995.
M. Verleysen, P. Thissen, J.-L. Voz, J. Madrenas, “An Analog Processor Architecture for a Neural Network Classifier”, IEEE MICRO, Vol. 14, No. 3, pp. 16–28, June 1994.
Thissen Ph., Verleysen M., Legat J.D., Madrenas J., Domínguez J., “A VLSI System for Neural Bayesian and LVQ Classification”, From Natural to Artificial Neural Computation, Editors: J. Mira, F. Sandoval, pp. 696–703, Springer-Verlag, 1995.
Denyer P.D., Renshaw D., VLSI Signal Processing: a Bit-Serial Approach, Addison-Wesley, 1985.
Thissen P., Verleysen M., Voz J.L., “Accuracy for Digital and Analog Implementation of Kernel Classifiers”, ESPRIT Project ELENA-Nerves 2 (no. 6891) Deliverable R2-C-P, June 1994.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Madrenas, J., Ruiz, G., Moreno, J.M., Cabestany, J. (1997). Synthesis and optimization of a bit-serial pipeline kernel processor. In: Mira, J., Moreno-Díaz, R., Cabestany, J. (eds) Biological and Artificial Computation: From Neuroscience to Technology. IWANN 1997. Lecture Notes in Computer Science, vol 1240. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0032539
Download citation
DOI: https://doi.org/10.1007/BFb0032539
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63047-0
Online ISBN: 978-3-540-69074-0
eBook Packages: Springer Book Archive