A programmable cellular neural network circuit

Author

Leong, Michel ; Vasconcelos, Pedro ; Fernandes, Jorge R. ; Sousa, Leonel

Author_Institution

INESC, Instituto Superior Tecnico, Lisbon, Portugal

fYear

2004

fDate

7-11 Sept. 2004

Firstpage

186

Lastpage

191

Abstract

In this paper, we propose and develop a fully programmable CNN circuit. The CNN coefficients are digitally programmable using a digital to analog converter (DAC), resulting in added flexibility. CNNs with 4×4 and 16×16 cells are designed and tested, exhibiting good accuracy when compared with Matlab and Java applications for computing CNNs. All circuits are designed and implemented with a 0.35 μm CMOS technology. The layout of a full 4×4 CNN was designed using cadence design framework II. The circuits are simulated with Pspice/Spectre.

Keywords

SPICE; cellular neural nets; circuit simulation; digital-analogue conversion; integrated circuit layout; integrated circuit testing; 0.35 micron; CMOS technology; DAC; Java; Matlab; Pspice/Spectre; cadence design framework; digital-analog converter; programmable cellular neural network circuit; Algorithm design and analysis; CMOS technology; Capacitors; Cellular neural networks; Integrated circuit interconnections; Multidimensional signal processing; Permission; Signal processing; Very large scale integration; Voltage;

fLanguage

English

Publisher

ieee

Conference_Titel

Integrated Circuits and Systems Design, 2004. SBCCI 2004. 17th Symposium on

Print_ISBN

1-58113-947-0

Type

conf

DOI

10.1109/SBCCI.2004.240873

Filename

1360567