Traffic Matrix Estimation Using The Levenberg- Marquardt Neural Network Of A Large IP System
Free (open access)
85 - 95
S. Mekaoui1, C. Benhamed1 & K. Ghoumid
This paper deals with a method using a specific class of neural networks whose learning phase is based on the Levenberg-Marquardt algorithm and which had been applied to the estimation of the traffic matrix (TM) of a large scale IP network. The neural network had been implemented with the help of the specific neural toolbox of the source software Matlab. Such neural networks are within the class of feed forward and recurrent types. The simulation tests have been processed on the available data base of the very reputed American observatory data base on the Internet of a very large scale IP network, the so-called Abilene network on both categories of neural networks. The simulated results using this method have been found to be very accurate as compared to one another. The static model converges rapidly but was less accurate in the estimation of the Traffic Matrix of such a kind of large IP System (the Abilene System) than the dynamic model which in this way earned the challenge of yielding a perfect estimation. Keywords: IP networks, Traffic Matrix (TM), neural networks (NN), Levenberg- Marquardt learning algorithm, Traffic Matrix estimation, linear regression.
Keywords: IP networks, Traffic Matrix (TM), neural networks (NN), Levenberg- Marquardt learning algorithm, Traffic Matrix estimation, linear regression.