Scheduling of routing table calculation schemes in open shortest path first using artificial neural network

Abstract

Internet topology changes due to events such as router or link goes up and down. Topology changes trigger routing protocol to undergo convergence process which eventually prepares new shortest routes needed for packet delivery. Real-time applications (e.g. VoIP) are increasingly being deployed in internet nowadays and require the routing protocols to have quick convergence times in the range of milliseconds. To speed-up its convergence time and better serve real-time applications, a new routing table calculation scheduling schemes for Interior Gateway Routing Protocol called Open Shortest Path First (OSPF) is proposed in this research. The proposed scheme optimizes the scheduling of OSPF routing table calculations using Artificial Neural Network technique called Generalized Regression Neural Network. The scheme determines the suitable hold time based on three parameters: LSA-inter arrival time, the number of important control message in queue, and the computing utilization of the routers. The GRNN scheme is tested using Scalable Simulation Framework (SSFNet version 2.0) network simulator. Two kind of network topology with several link down scenarios used to test GRNN scheme and existing scheme (fixed hold time scheme). Results shows that GRNN provide faster convergence time compared to the existing scheme.