DocumentCode :
2947120
Title :
Implementations of Hopfield neural network in communication networks
Author :
Kojic, Nenad S.
Author_Institution :
ICT Coll. of vocational studies, Belgrade, Serbia
fYear :
2013
fDate :
26-28 Nov. 2013
Firstpage :
397
Lastpage :
404
Abstract :
Hopfield neural network, as a recurrent neural network, shows very good results in solving a lot of different complex computational problems. Starting from its previous modification, and making the new ones, this network was implemented in different types of communication networks in order to solve some real problems. In this paper, the possibility of intelligent decision making of Hopfield neural network, through the three independent implementations, will be presented. Starting from the problem of dynamic routing, one possible solution for multicast routing in telecommunication networks as well as routing in all optical networks will be presented. Beside this, the modifications of Hopfield neural network could be used for planning and projection of the new network infrastructures. One solution for route selection problem in real physical environment will be shown.
Keywords :
Hopfield neural nets; decision making; multicast communication; optical fibre networks; telecommunication computing; telecommunication network planning; telecommunication network routing; Hopfield neural network; all optical networks; communication networks; dynamic routing; intelligent decision making; multicast routing; network infrastructures; recurrent neural network; route selection problem; telecommunication networks; Cities and towns; Heuristic algorithms; Hopfield neural networks; Network topology; Neurons; Optical wavelength conversion; Routing; Hopfield neural network; Multicast routing; Route selection problem; Routing algorithm; Shortest path;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Telecommunications Forum (TELFOR), 2013 21st
Conference_Location :
Belgrade
Print_ISBN :
978-1-4799-1419-7
Type :
conf
DOI :
10.1109/TELFOR.2013.6716253
Filename :
6716253
Link To Document :
بازگشت