Home Author Index Search Volume 1 May 2009 ISSN 1797-9617

International Journal of

Recent Trends in Engineering

Home > Vol. 1, No. 1


International Journal of Recent Trends in Engineering (IJRTE)

ISSN 1797-9617

Volume 1, Number 1, May 2009

Issue on Computer Science

Page(s): 387-392

Neural Network Based Switching Models for VLSI Design of ATM Networks

R.Vijayakumar, Sushama. G

Full text: PDF


In this paper we present a framework for designing neural network solutions using extensions of the competitive learning Artificial Neural Network(ANN) model for ATM networks in VLSI design. Unlike conventional LANs ,ATM LANs utilize a mesh topology comprising a number of interconnected switching exchanges similar in principle to that used in conventional telephone networks. ATM switch has a defined number of ports and its function is to provide a high bit rate switched communications paths between ports. The complexity of switching can be reduced by introducing the ANN based massively parallel and distributed switching topology for VLSI . ANN based networks have properties that can be directly defined and controlled. This direct approach allows efficient implementations that are scalable to large sizes and as long as the external inputs are within defined limits. The network will always satisfy the constraints embodied in the competitive learning models. This is an efficient means for communicating the many constraints of the problems between the neurons in our solutions. The neural network designs based on simple electrical components have a direct implementation in hardware. We had designed a multiple overlapping competitive learning network model. This paper concentrate on the new technique of teaching of ANNs through a method called PERTURBATE - COMPARE - AND - CORRECTION method ,which is a more analog approach to teaching an ANN and implementation in Large scale integration Technology. But it can also be used for digital applications . Basically the matrix model is chosen for the time being for the neural network. With this the teaching circuit, which works in a multiplexed mode, can access each weight and update them in a systematic manner. The circuit was build using spice simulator and simulation works were completed successfully for Large Scale integration models.

Index Terms


Published by Academy Publisher in cooperation with the ACEEE

@ Copyright 2009 ACADEMY PUBLISHER All rights reserved