Title :
Hopfield-genetic approach for solving the routing problem in computer networks
Author :
Hamdan, M. ; El-Hawary, M.E.
Author_Institution :
Electr. & Comput. Eng. Dept., Dalhousie Univ., Halifax, NS, Canada
Abstract :
This paper presents a method that combines Hopfield networks (HN) and a genetic algorithm (GA) to solve the problem of optimal routing in computer networks. Hopfield neural networks perform fast and efficient local searches and guarantee convergence to feasible solutions, but they are inefficient for global search. On the other hand, genetic algorithms are powerful in global search, but they take long computation time to solve large-scale problem. To overcome these limitations, a combined approach is proposed where a solution produced by a Hopfield network is incorporated in the initial population of the GA. The proposed method is verified through simulation and it shows an improvement in the quality of the solution and reduces the computation time.
Keywords :
Hopfield neural nets; computer networks; genetic algorithms; minimisation; network topology; search problems; telecommunication network routing; Hopfield neural networks; computation time; computer networks; genetic algorithm; global search; local search; minimisation; network topology; packet switching networks; routing problem; Computational modeling; Computer networks; Costs; Genetic algorithms; Intelligent networks; Large-scale systems; Network topology; Neural networks; Routing; Spread spectrum communication;
Conference_Titel :
Electrical and Computer Engineering, 2002. IEEE CCECE 2002. Canadian Conference on
Print_ISBN :
0-7803-7514-9
DOI :
10.1109/CCECE.2002.1013048
