Title :
A charge-based on-chip adaptation Kohonen neural network
Author :
Y. He;U. Cilingiroglu
Author_Institution :
Dept. of Electr. Eng., Texas A&M Univ., College Station, TX, USA
Abstract :
A charge-based on-chip synapse adaptation Kohonen neural network circuit is proposed. The properties of the approach are low power dissipation and high density due to the charge transfer mechanism and the novel compact device configurations. The prototype chip which contains 12*10 synapses with a density of 190 synapses/mm/sup 2/ was fabricated with 2- mu m standard CMOS technology. The experimental results from the prototype chip demonstrated successful unsupervised learning and classification as theoretically predicted.
Keywords :
"Network-on-a-chip","Neural networks","Circuits","Adaptive systems","Mechanical factors","Power dissipation","Charge transfer","Prototypes","CMOS technology"
Journal_Title :
IEEE Transactions on Neural Networks