An artificial neural network approach for routing in distributed computer networks

This paper presents a new approach based on the Hopfield model of artificia l neural networks to solve the routing problem in a context of computer net work design. The computer networks considered are packet switching networks . modeled as non-oriented graphs where nodes represent servers, hosts or sw itches, while bi-directional and symmetric arcs represent full duplex commu nication links. The proposed method is based on a network representation en abling to match each network configuration with a Hopfield neural network i n order to find the best path between any node pair by minimizing an energy function. The results show that the time delay derived from Row assignment carried out by this approach is, in most cases, better than those determin ed using conventional routing heuristics. Therefore, this neural-network ap proach is suitable to be integrated into an overall topological design proc ess of moderate-speed and high-speed networks subject to quality of service constraints as well as to changes in configuration and link costs. (C) 200 1 Published by Elsevier Science Ltd.