Title :
Low-power CMOS circuits for analog VLSI programmable neural networks
Author :
El-soud, Mohy A. ; AbdelRassoul, Roshdy A. ; Soliman, Hassan H. ; El-ghanam, Laila M.
Author_Institution :
Dept. of Electron. & Commun. Eng., Mansoura Univ., Egypt
Abstract :
This paper presents an analog VLSI neural network for designing a programmable neural system. Synaptic weights are designed in the triode region using four-MOS transistors. Moreover, the summing element (SE) and the activation function are designed in subthreshold region. This system is realized in a standard 0.8 μm CMOS technology and operated with a ±1V power supply.
Keywords :
CMOS integrated circuits; MOSFET; VLSI; neural nets; semiconductor device models; summing circuits; 0.8 micron; CMOS technology; MOS transistors; analog VLSI programmable neural networks; low-power CMOS circuits; subthreshold region; summing element; Artificial neural networks; CMOS analog integrated circuits; CMOS technology; Handwriting recognition; MOSFETs; Neural networks; Neurons; Strontium; Threshold voltage; Very large scale integration;
Conference_Titel :
Microelectronics, 2003. ICM 2003. Proceedings of the 15th International Conference on
Print_ISBN :
977-05-2010-1
DOI :
10.1109/ICM.2003.1287711
