Title :
An MOS cell circuit for compact implementation of reaction-diffusion models
Author :
K. Karahaliloglu;S. Balkir
Author_Institution :
Dept. of Electr. Eng., Nebraska-Lincoln Univ., Lincoln, NE, USA
fDate :
6/26/1905 12:00:00 AM
Abstract :
A bio-inspired MOS cell design, which can be used in reaction-diffusion systems, and the corresponding implementation of a test reaction-diffusion CNN (RD-CNN) layer is presented. The network architecture and an approximate analysis of its certain D.C. characteristics are given. A test chip, which includes an array of the connected cells, is implemented. The chip also has separate test cells, which employ thick oxide MOSFET. The wave propagation in the network and I-V characteristics of the cells using thick-oxide MOSFETs are demonstrated and verified with measured results. The cell circuits and the resulting network are very compact and easily implementable with available technologies.
Keywords :
"Cellular neural networks","CMOS technology","Biological system modeling","MOSFETs","Laplace equations","Circuit testing","Cells (biology)","Semiconductor device modeling","Buildings","Design engineering"
Conference_Titel :
Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on
Print_ISBN :
0-7803-8359-1
DOI :
10.1109/IJCNN.2004.1379886