In search of a robust digital CNN system

Wen Hai Fang; Cheng Wang; Lambert Spaanenburg

doi:10.1109/CNNA.2006.341654

In search of a robust digital CNN system

Wen Hai Fang, Cheng Wang, Lambert Spaanenburg

Department of Electrical and Information Technology

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

Abstract

Microelectronics promises a high component density and low power dissipation to embedded systems. Unfortunately such a component will always suffer from various error types that make the chip respond differently from its functional simulation. This is especially true for Cellular Neural Networks (CNN), which makes the determination of robust, low-precision parameters to guarantee small footprint and reliable operation an important design consideration. This paper describes the digital word width effects in a CNN implementation that must be considered to achieve a small size for a reliable system. It discusses the automated design space exploration using a Field-Programmable Gate-Array (FPGA) implementation to perform an optimal CNN parameters selection.

Original language	English
Title of host publication	Proceedings of the 2006 10th IEEE International Workshop on Cellular Neural Networks and Their Applications
Publisher	IEEE - Institute of Electrical and Electronics Engineers Inc.
Pages	328-333
Number of pages	6
ISBN (Print)	1-4244-0639-0
DOIs	https://doi.org/10.1109/CNNA.2006.341654
Publication status	Published - 2006
Event	10th International Workshop on Cellular Neural Networks and their Applications (CNNA) - Istanbul Duration: 2006 Aug 28 → 2006 Aug 30

Conference

Conference	10th International Workshop on Cellular Neural Networks and their Applications (CNNA)
Period	2006/08/28 → 2006/08/30

Subject classification (UKÄ)

Electrical Engineering, Electronic Engineering, Information Engineering

Free keywords

automated design space exploration
digital word width effects
cellular neural networks
microelectronics
robust templates
embedded systems
field-programmable gate-array

Access to Document

10.1109/CNNA.2006.341654

Cite this

@inproceedings{ac2bd5b062264b8cbd01099d537be380,

title = "In search of a robust digital CNN system",

abstract = "Microelectronics promises a high component density and low power dissipation to embedded systems. Unfortunately such a component will always suffer from various error types that make the chip respond differently from its functional simulation. This is especially true for Cellular Neural Networks (CNN), which makes the determination of robust, low-precision parameters to guarantee small footprint and reliable operation an important design consideration. This paper describes the digital word width effects in a CNN implementation that must be considered to achieve a small size for a reliable system. It discusses the automated design space exploration using a Field-Programmable Gate-Array (FPGA) implementation to perform an optimal CNN parameters selection.",

keywords = "automated design space exploration, digital word width effects, cellular neural networks, microelectronics, robust templates, embedded systems, field-programmable gate-array",

author = "Fang, {Wen Hai} and Cheng Wang and Lambert Spaanenburg",

year = "2006",

doi = "10.1109/CNNA.2006.341654",

language = "English",

isbn = "1-4244-0639-0",

pages = "328--333",

booktitle = "Proceedings of the 2006 10th IEEE International Workshop on Cellular Neural Networks and Their Applications",

publisher = "IEEE - Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "10th International Workshop on Cellular Neural Networks and their Applications (CNNA) ; Conference date: 28-08-2006 Through 30-08-2006",

}

Fang, WH, Wang, C & Spaanenburg, L 2006, In search of a robust digital CNN system. in Proceedings of the 2006 10th IEEE International Workshop on Cellular Neural Networks and Their Applications. IEEE - Institute of Electrical and Electronics Engineers Inc., pp. 328-333, 10th International Workshop on Cellular Neural Networks and their Applications (CNNA), 2006/08/28. https://doi.org/10.1109/CNNA.2006.341654

In search of a robust digital CNN system. / Fang, Wen Hai; Wang, Cheng; Spaanenburg, Lambert.
Proceedings of the 2006 10th IEEE International Workshop on Cellular Neural Networks and Their Applications. IEEE - Institute of Electrical and Electronics Engineers Inc., 2006. p. 328-333.

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

TY - GEN

T1 - In search of a robust digital CNN system

AU - Fang, Wen Hai

AU - Wang, Cheng

AU - Spaanenburg, Lambert

PY - 2006

Y1 - 2006

N2 - Microelectronics promises a high component density and low power dissipation to embedded systems. Unfortunately such a component will always suffer from various error types that make the chip respond differently from its functional simulation. This is especially true for Cellular Neural Networks (CNN), which makes the determination of robust, low-precision parameters to guarantee small footprint and reliable operation an important design consideration. This paper describes the digital word width effects in a CNN implementation that must be considered to achieve a small size for a reliable system. It discusses the automated design space exploration using a Field-Programmable Gate-Array (FPGA) implementation to perform an optimal CNN parameters selection.

AB - Microelectronics promises a high component density and low power dissipation to embedded systems. Unfortunately such a component will always suffer from various error types that make the chip respond differently from its functional simulation. This is especially true for Cellular Neural Networks (CNN), which makes the determination of robust, low-precision parameters to guarantee small footprint and reliable operation an important design consideration. This paper describes the digital word width effects in a CNN implementation that must be considered to achieve a small size for a reliable system. It discusses the automated design space exploration using a Field-Programmable Gate-Array (FPGA) implementation to perform an optimal CNN parameters selection.

KW - automated design space exploration

KW - digital word width effects

KW - cellular neural networks

KW - microelectronics

KW - robust templates

KW - embedded systems

KW - field-programmable gate-array

U2 - 10.1109/CNNA.2006.341654

DO - 10.1109/CNNA.2006.341654

M3 - Paper in conference proceeding

SN - 1-4244-0639-0

SP - 328

EP - 333

BT - Proceedings of the 2006 10th IEEE International Workshop on Cellular Neural Networks and Their Applications

PB - IEEE - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th International Workshop on Cellular Neural Networks and their Applications (CNNA)

Y2 - 28 August 2006 through 30 August 2006

ER -

In search of a robust digital CNN system

Abstract

Conference

Subject classification (UKÄ)

Free keywords

Access to Document

Fingerprint

Cite this