Neural Network Based Routing Algorithm for Cognitive Packet Network Architecture

Abstract

This journal focuses on introducing an adaptive routing algorithm to be used on Wide Area Network(WAN) like Internet, based on the Cognitive Packet Network (CPN) architecture to enhance the Quality of Service (QoS) delivered to the end users. This research focuses on Building on top of the CPN network architecture and introducing a neural network based routing algorithm which would improve the QoS, provide reliable and efficient service which can fulfill the ever growing Internet usage demand. The neural network’s task would be learning the network and adapt to any situation through the knowledge gathered. The packets will collectively learn about the network thus the load on the routers will be minimized. This mechanism completely replaces traditional method of routing tables which results in efficient and reliable packet delivery in comparison to current routing protocols like Open Shortest Path First (OSPF). This research has introduced a routing algorithm which is Intelligent, dynamically adapting and self-learning for CPN infrastructure instead of current packet switched network.