In this paper, flows classification in IP switch is discussed. Different traditional flow classifiers are briefly introduced. Then a Hopfield neural network flow classifier (HNNC) is proposed. It can detect services as suitable for switching in accordance with VC space of IP switch. The simulation results show that services suitable to be switched are detected by HNNC and the majority of packets are assigned to a minimum number of flows.
Index Terms:
IP switching, flow classification, neural network
Citation:
Lianyuan Li, Zemin Liu, Zheng Zhou, "A Hopfield Neural Network Flow Classifier in IP Switching," ijcnn, vol. 5, pp.5398, IEEE-INNS-ENNS International Joint Conference on Neural Networks (IJCNN'00)-Volume 5, 2000