Programmable gm-C CNN implementation

Drahoslav Lim; George S. Moschytz

Programmable g_m-C CNN implementation

Drahoslav Lim, George S. Moschytz

Swiss Federal Institute of Technology Zurich

Research output: Contribution to conference › Paper › peer-review

2 Scopus citations

Abstract

An implementation of a programmable cellular neural network is reported. It overcomes some of the limiting characteristics and restrictions inherent in CMOS VLSI technologies, and allows an arbitrarily large continuous-time analog CNN to be built up by modularly connecting CNN chips with a modest number of cells. The template values are implemented as sets of unit and half-unit OTAs and are digitally step-wise programmable. The design incorporates an offset compensation and initialization circuit. All external input, output and control signals are electrical and digital. The design was carried out in a 0.8 μ CMOS technology. Each cell occupies 0.78 mm², including all support circuitry. Matching accuracy was measured and operation was verified on numerous uncoupled and propagation-type templates.

Original language	English
Pages	294-299
Number of pages	6
State	Published - 1998
Externally published	Yes
Event	Proceedings of the 1998 5th IEEE International Workshop on Cellular Neural Networks and Their Applications, CNNA - London, UK Duration: 14 Apr 1998 → 17 Apr 1998

Conference

Conference	Proceedings of the 1998 5th IEEE International Workshop on Cellular Neural Networks and Their Applications, CNNA
City	London, UK
Period	14/04/98 → 17/04/98

Cite this

@conference{54668a2114f84e54b9200257df3a7712,

title = "Programmable gm-C CNN implementation",

abstract = "An implementation of a programmable cellular neural network is reported. It overcomes some of the limiting characteristics and restrictions inherent in CMOS VLSI technologies, and allows an arbitrarily large continuous-time analog CNN to be built up by modularly connecting CNN chips with a modest number of cells. The template values are implemented as sets of unit and half-unit OTAs and are digitally step-wise programmable. The design incorporates an offset compensation and initialization circuit. All external input, output and control signals are electrical and digital. The design was carried out in a 0.8 μ CMOS technology. Each cell occupies 0.78 mm2, including all support circuitry. Matching accuracy was measured and operation was verified on numerous uncoupled and propagation-type templates.",

author = "Drahoslav Lim and Moschytz, {George S.}",

year = "1998",

language = "אנגלית",

pages = "294--299",

note = "Proceedings of the 1998 5th IEEE International Workshop on Cellular Neural Networks and Their Applications, CNNA ; Conference date: 14-04-1998 Through 17-04-1998",

}

TY - CONF

T1 - Programmable gm-C CNN implementation

AU - Lim, Drahoslav

AU - Moschytz, George S.

PY - 1998

Y1 - 1998

N2 - An implementation of a programmable cellular neural network is reported. It overcomes some of the limiting characteristics and restrictions inherent in CMOS VLSI technologies, and allows an arbitrarily large continuous-time analog CNN to be built up by modularly connecting CNN chips with a modest number of cells. The template values are implemented as sets of unit and half-unit OTAs and are digitally step-wise programmable. The design incorporates an offset compensation and initialization circuit. All external input, output and control signals are electrical and digital. The design was carried out in a 0.8 μ CMOS technology. Each cell occupies 0.78 mm2, including all support circuitry. Matching accuracy was measured and operation was verified on numerous uncoupled and propagation-type templates.

AB - An implementation of a programmable cellular neural network is reported. It overcomes some of the limiting characteristics and restrictions inherent in CMOS VLSI technologies, and allows an arbitrarily large continuous-time analog CNN to be built up by modularly connecting CNN chips with a modest number of cells. The template values are implemented as sets of unit and half-unit OTAs and are digitally step-wise programmable. The design incorporates an offset compensation and initialization circuit. All external input, output and control signals are electrical and digital. The design was carried out in a 0.8 μ CMOS technology. Each cell occupies 0.78 mm2, including all support circuitry. Matching accuracy was measured and operation was verified on numerous uncoupled and propagation-type templates.

UR - http://www.scopus.com/inward/record.url?scp=0031635257&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontoconference.paper???

AN - SCOPUS:0031635257

SP - 294

EP - 299

T2 - Proceedings of the 1998 5th IEEE International Workshop on Cellular Neural Networks and Their Applications, CNNA

Y2 - 14 April 1998 through 17 April 1998

ER -

Programmable g_m-C CNN implementation

Abstract

Conference

Other files and links

Fingerprint

Cite this