Fair bandwidth control in ATM-systems by artificial neural networks

This paper is part of a research project to examine the use of fuzzy controllers and artificial neural networks in the resource management of high speed networks. It deals with an investigation on adaptive flow-control mechanism based on artificial neural networks integrated in ATM networks. In addition to a new overall concept, two new methods for static and dynamic assignment of priorities for individual services or connections are presented. The results are verified by means of extensive simulations.