Title :
A new approach to design cellular neural networks for associative memories
Author :
Grassi, Giuseppe
Author_Institution :
Dipt. di Matematica, Lecce Univ., Italy
fDate :
9/1/1997 12:00:00 AM
Abstract :
In this brief, a synthesis procedure of cellular neural networks for associative memories is presented, The proposed method, by assuring the global asymptotic stability of the equilibrium point, generates networks where the input data are fed via external inputs rather than initial conditions. This new approach enables to design both heteroassociative and autoassociative memories and reveals particularly suitable for VLSI implementation techniques
Keywords :
asymptotic stability; cellular neural nets; circuit stability; content-addressable storage; integrated circuit design; neural chips; VLSI implementation; associative memories; autoassociative memories; cellular neural network design; equilibrium point; external inputs; global asymptotic stability; heteroassociative memories; synthesis procedure; Associative memory; Asymptotic stability; Cellular neural networks; Circuits; Design methodology; Equations; Network synthesis; Steady-state; Turing machines; Very large scale integration;
Journal_Title :
Circuits and Systems I: Fundamental Theory and Applications, IEEE Transactions on
