Adaptive ant-based dynamic routing algorithm

Abstract

Network topologies are not only continuously growing, but many devices become mobile and some sort of guaranteed quality of service (QoS) is now required. In order to efficiently deal with the changing traffic loads and topologies, reduce the phenomena of congestions, this paper describes a new adaptive dynamic routing algorithm for packet-switched communications networks based on simple biological "ants" that explore the network and learn good routes, using a novel variation of reinforcement learning. Simulation results confirm that the algorithm is shown to significantly improve the network's relaxation and its response to perturbations.