Hopfield Neural Network for Routing

In the context of the Packet Switching Networks, the main need is to optimize the paths connecting the source and the destination. In this paper the authors deal with the problem of routing by means of a neural approach. This is based on the assumption that each switching node is made up by a Hopfield network which is capable to determine the optimal path which connects the local node with a generic node in the packet switching network. On this basis, the suitable output link on which each incoming packet must be routed is chosen.

[1]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[2]  Edsger W. Dijkstra,et al.  A note on two problems in connexion with graphs , 1959, Numerische Mathematik.

[3]  S. Cavalieri,et al.  A neural-network-based approach for routing in a packet switching network , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.