Neural Networks for Shortest Path Computation and Routing in Computer Networks

Many intellectual optimization performances like Artificial Neural Networks (ANN),Genetic Algorithms (GAs), etc., were being planned to find the stagnant shortest path. Rapid expansions in the wireless communication predominantly in the field of mobile networks havematerialized as two major fields namely Mobile Ad hoc Networks (MANETs) and Wireless Sensor Networks (WSN). Topology elusiveness is the top most challenge in the mobile wireless network field i.e., the network topology changes over time due to energy conservancy or node mobility. In order to find the shortest path (SP) with in this network becomes a dynamic optimization problem due to nodes mobility. Nodes usually die due to low energy or it may move, this scenario makes the network to be more complex for finding shortest path. In this paper we propose a novel method of using Genetic Algorithms (GAs) to solve the dynamic shortest path discovery and routing in MANETs. MANETs is one of the faster growing new-generation wireless networks. The tentative results indicate that this GA based algorithm can quick adapt to environmental change (i.e. the network topology change) and create high quality solutions after each change.