A Neural-Net Based Fuzzy Admission Controller for an ATM Network

This paper pro oses a neural fuzzy connection admission control NFCAC) scheme, which combines of the neural-net, t o solve the connection admission control (CAC) problems in ATM networks. Recently, fuzzy logic systems have been successfully applied t o deal with the traffic control related probleims and provided a robust mathematical framework for dealing with "real-world" imprecision; multi-layer neural networks are capable of producing complex de:cisions with arbitrarily nonlinear boundaries and the:y have been used as solution for the CAC. However, the application of neural network or fuzzy logic sysliem to CAC exists some difficulties in real operation. The proposed NFCAC solves the difficulties b y combini*ng the benefits of the existing traffic control mechanicims, linguistic control strategy of the fuzzy logic controller and the learning ability of neural-net. Simuldion results show that the proposed NFCAC saves a large amount of training time and simplifies the design procedure of a CAC controller but provides a superior system utilization, while keeping the QoS contract, than either neural network or fuzzy logic system does. benefits of fuzzy I p ogic controller and learning abilities